Higher Education, Development, and Inequality in Brazil and South Africa

This article has the premise that South Africa and Brazil spaces share contextual and geopolitical characteristics with a history of great inequalities, racial and gender discrimination and these and other related factors serve as barriers constraining education. Considering the remarkable expansion of higher education systems in both countries on the last 25 years, and its uneven effects, some questions are raised as a challenge in this article. Does this growth in enrolments create high quality or “world class universities” in these countries? Is it possible to find South African or Brazilian universities in the international rankings of institutional higher education? Has such expansion produced a full democratization of educational opportunities? Or, in other words, does any skilled and hardworking student, regardless of his/her social background, have equal chances of access to the best courses and universities? In order to try to answer these questions, we begin characterizing the expansion of higher education systems over the last two and a half decades in both countries. Regarding policies of access by poor students to higher education system, we taking in account and compare some initiatives in both countries, such as Reuni, Fies and Prouni in Brazil, and National Student Financial Aid Scheme (NSFAS), in South Africa. Our analysis, following the tradition of sociological research, understands that the mode of operation of higher education Changing Societies & Personalities, 2018, Vol. 2, No. 4, pp. 366–392 367 Introduction: Higher Education in Brazil and South Africa For two and a half decades the Brazilian system of higher education (HES) has expanded and diversified. New courses and new school paths were offered as well as different types of diplomas. Brazil went from 1.3 million students in 1980 to 8 million in 2015. Hundreds of thousands from working and so-called “popular classes” entered the university. Public policies such as quotas or scholarships ensured a significant expansion of access to tertiary education. In South Africa, in 1993 – before the end of Apartheid – there were 473,000 students HE students enrolled, this grew to 683,000 in 1996 and reached 2,000,000 in 2016. South Africa also moved to “promote equity of access and fair chances for success to all who are seeking to realize their potential through higher education, while eradicating all forms of unfair discrimination and advancing redress for past inequalities” (White Paper, 1997, 1.14) Does this growth in enrolments create high quality or “world class universities” in these countries? Is it possible to find South African or Brazilian universities in the international rankings of institutional higher education? Has such expansion produced a full democratization of educational opportunities? Or, in other words, does any skilled and hardworking student, regardless of his/her social background, have equal chances of access to the best courses and universities? The last is the key question for the sociology of social inequality: does expansion “reduce inequality by providing more opportunities for persons from disadvantage strata, or magnify inequality by expanding opportunities disproportionately for those who are already privileged”? (Arum et al., 2007, p. 1) The production of quality rankings includes, at least in countries like the USA, measures of social, gender and racial inequalities but it is mainly associated with scientific research and learning (Goastellec, 2008). For analysts of most diverse hues, the proper functioning of institutions of higher education – their economic, social, and scientific efficiency – has become an important element in the assertion of democratic principles and equality of institutions stands out as one of the key factors in the mechanisms and social conflicts that increase or reduce inequalities. Focusing on the basic distinction between public and private sector, for Brazil, and the persistence of distinction between historically black and white institutions, in South Africa, we try to show that both countries improved the access to higher education systems and managed to create some worldclass institutions. Even so, social and gender inequalities persist and there are too few such institutions, especially in Brazil.


Introduction: Higher Education in Brazil and South Africa
For two and a half decades the Brazilian system of higher education (HES) has expanded and diversified. New courses and new school paths were offered as well as different types of diplomas. Brazil went from 1.3 million students in 1980 to 8 million in 2015. Hundreds of thousands from working and so-called "popular classes" entered the university. Public policies such as quotas or scholarships ensured a significant expansion of access to tertiary education. In South Africa, in 1993 -before the end of Apartheid -there were 473,000 students HE students enrolled, this grew to 683,000 in 1996 and reached 2,000,000 in 2016. South Africa also moved to "promote equity of access and fair chances for success to all who are seeking to realize their potential through higher education, while eradicating all forms of unfair discrimination and advancing redress for past inequalities" (White Paper, 1997, 1.14) Does this growth in enrolments create high quality or "world class universities" in these countries? Is it possible to find South African or Brazilian universities in the international rankings of institutional higher education? Has such expansion produced a full democratization of educational opportunities? Or, in other words, does any skilled and hardworking student, regardless of his/her social background, have equal chances of access to the best courses and universities? The last is the key question for the sociology of social inequality: does expansion "reduce inequality by providing more opportunities for persons from disadvantage strata, or magnify inequality by expanding opportunities disproportionately for those who are already privileged"? (Arum et al., 2007, p. 1) The production of quality rankings includes, at least in countries like the USA, measures of social, gender and racial inequalities but it is mainly associated with scientific research and learning (Goastellec, 2008).
For analysts of most diverse hues, the proper functioning of institutions of higher education -their economic, social, and scientific efficiency -has become an important element in the assertion of democratic principles and equality of institutions stands out as one of the key factors in the mechanisms and social conflicts that increase or reduce inequalities. Focusing on the basic distinction between public and private sector, for Brazil, and the persistence of distinction between historically black and white institutions, in South Africa, we try to show that both countries improved the access to higher education systems and managed to create some world-class institutions. Even so, social and gender inequalities persist and there are too few such institutions, especially in Brazil.

KEYWORDS
Brazil, South Africa, higher education, affirmative action, education inequalities opportunities in modern societies. This understanding seems to be present in Brazilian and South African societies even if diverse social groups have different expectations and demands for higher education. These groups have political strategies and ways to lobby the State and society to try to embed the models they think suitable for themselves and for all in the education system. However, the common perception that some level of democratization is necessary in higher education emerges through the multiplication of public and institutional policies for the inclusion of previously excluded social groups.
Numerous student assistance policies have been developed in Brazilian public universities to facilitate some kind of integration of previously excluded students (Heringer & Honorato, 2014, p. 323). Some initiatives created in private institutionsscholarship programs and internships -can be considered as evidence of this increasingly widespread democratizing impetus (Almeida Neto, 2015, p. 23). Several affirmative action policies have been developed in Brazil since the 1990s, with some success in the inclusion of afro-descendants and poor youth. More recently, REUNI (Restructuring and Expansion of Federal Universities Program) promoted a significant increase in the number of federal institutions of higher education and the increased the offer of student vacancies outside of the populated coastal areas. In a continental country such as Brazil, this program, despite its difficulties, greatly facilitates the access of young people with limited resources to go to good universities near their homes (Vargas, 2014). Similarly, the federal government's PROUNI (University for All Program) provided grants and scholarships for poor students to study in private universities (Santos, 2012). The federal government also invested heavily in vocational centers at the tertiary level. This aimed to increase the alternatives to university education, thus differing from the two aforementioned policies (Mont'Alvão, 2015).
In South Africa, policies try to interweave economic development, a greater equality of opportunities and the overcoming of past inequalities. The increase of enrollments in South Africa's higher education system did not begin in 1994, but it was already present during the apartheid period. In fact, as shown by Akoojee & Nkomo (2008), the demand for skilled people in apartheid times produced a dual higher education system in the country. In the one hand, a university for the white population, focused on scientific knowledge to foster the ruling class. On the other hand, a university for Indians and Coloreds, focused on training for the labor market. Even though a large mass of the population was denied access to higher education, "economic imperatives under apartheid left some room for selected black people to access institutions of higher education" (Akoojee & Nkomo, 2008, p. 389). Some authors (Patto, 2007) (Libâneo, 2012 have argued that a similar dual education system was built in Brazil, without such emphasis on racial and ethnic discrimination, but which stressed the separation between poor and rich students.
In the South Africa's transition from apartheid, when the need to redress past inequalities became a priority, a political agenda that increased access for black and colored groups to the higher education system was developed. As pointed out by Akoojee and Nkomo: Transformation requires that the ethos that prevailed at higher education institutions (HEIs) in the past needs to be replaced with a new democratic culture directed at actively undoing race-based separation. In this regard, the issue of access to higher education institutions remains the key mechanism by which to forge a new order (Akoojee & Nkomo, 2008, p. 390).
Several changes in the political framework post-1994, such as the South Africa Constitution in 1996, the 1997 Act and the publication of a White Paper, produced deep alterations throughout the Higher Education System (Badat, 2010). Jansen (2007) sumarizes five major changes in the higher education landscape. First, a programme of government mergers in order to reduce the number of post-high school public institutions (numbering 306 these were radically reduced to 72 institutions). Second, a "spectacular growth in private higher education" (Jansen, 2007, p. 164) strengthened this segment in political and economic terms. The emergence of new models of delivering higher education was the third major change. So, according to (Jansen, 2007, p. 164), "it is no longer possible to clearly distinguish contact and distance education institutions in South Africa, as the former increasingly blurred the distinction in practice between these two forms of education delivery". The fourth change has been a decline in humanities enrolments. The last change was related to the nature of academic workplace, represented by the growth of a new managerialism characterized by "a growing emphasis on performance, measurement and accountability; the increasing ethos of competition; a changing language that recasts students as clients and departments as cost centres; the growing vulnerability of academic and administrative positions as 'outsourcing' and 'efficiencies' dominate the institutional strategy" (Jansen, 2007, p. 164).
The shift in the core rules of higher education system produces some uneven effects. One was the expansion of students enrolments. In 25 years enrollment almost doubled, from 473,000 in 1993 to some 800,000 in 2008 (Badat, 2010) . By 2016, 2 milion 1 students were enrolled in higher education, 78.3% of then in public institutions and 21.7% in private institutions (Statistics South Africa, 2017, p. 71). However, there are many discrepancies when taking into account ethnic and racial issues. In 2016, black and coloured people were expected to receive 1.7 years of schooling in higher education, whereas whites were expected to receive 7.5 years.
Since 1994, two kinds of policies of access for black students could be seen. First, "access as participation approach" (Akoojee & Nkomo, 2008, p. 390) when policies emphasize their increasing participation at universities that had previously denied them entry. Because of these efforts, Cloete and Bunting (2000) and Subotzky (2003) showed that the proportion of African students in White institutions increased, between 1993 and 2000, from 13% to 46%. Despite favorable enrollment growth numbers, there is evidence that this process did not guarantee success for the black students. A report quoted by Akoojee & Nkomo (2008, p. 390) indicated that at least 25 per cent of South Africa's higher education students fail to complete their studies. In addition, there is little black enrolment and success in high demand courses, such as science, engineering, and technology.
After 1999, "access with success" approach, which gives emphasis to guaranteeing success for these students was developed (Akoojee & Nkomo, 2008). In other words, these policies aim to provide proper conditions for black/poor students to successfully complete higher education. National Student Financial Aid Scheme (NSFAS), created in 1999, is an example of this kind of policy. The main purpose of the program is to enable young people from poor households to obtain a higher education. NSFAS provides loan and bursaries for students "access to, and success in, higher and further education and training" (National Student Financial Aid Scheme, 2018). In some dimensions, NFAS is quite similar to FIES in Brazil. Both provide loans to student and it is expected that they will be repaid when the new graduates enter the labor market. But there are some important differences. First, the loans could be used not only for undergraduate courses but also in some selected postgraduate programs 2 . The second and most important difference is that up to 40% of the loan can be converted into a bursary in South Africa, based on academic performance. A student led protest movement that began in 2015 called #FeesMustFall, in response to an increase in fees at South African universities. As a result, the 2018 NSFAS 3 will no longer disburse loans to students, according to South African President Jacob Zuma's December 2017 statement. It will only provide bursaries (Department of Higher Education and Training, 2018) NSFAS became more similar to the Brazilian Prouni than to FIES. Up until 2017, allowances for books, food, private accommodation, transport, and the like are paid directly to students using a voucher system (Government Technical Advisory Centre, 2018). In this sense, NSFAS faces a challenge that invisible in Brazilian FIES or Prouni, which is to provide adequate conditions so as to allow people to complete their courses successfully. In 2014, loans and grants were disbursed to about 425,000 students, twice as many as in 2010 (Government Technical Advisory Centre, 2018). In 2016, the NSFAS fund intended to support 405,000 first generation students in higher education and to continue to support those already in the system. (Statistics South Africa, 2017, p. 55). However, as we shall see below, some discrepancies surround South Africa government data. Based on information from Department of Higher Education and Training, 225,950 students were supported by NSFSAS fund in 2016.
The numbers in Graph 1 show that, in a broader perspective, both programs have exhibited a tendency of increase over the past decade4. The size of both programs, regarding the total number of students, is quite similar as well. But, as noted before, Brazilian´s Prouni is one of a number of programs, which include FIES and REUNI, that intend to increase the enrollment of poor students in higher education. Besides, Prouni is focused only on private education institutions.
South Africa's NSFSA and the Brazilian Prouni have one other similarity, which is the goal to promote teaching as a qualified profession. Part of the funds of NSFSA, for example, are addressed the Funza Lushaka bursary programme launched in 2007, which intends to improve the attractiveness of teaching as a career choice for South African students. The bursary provides fixed values per Higher Education Institution which should cover: a) student tuition fees, including laboratory fees (where applicable); b) residence/accommodation fees including meals; c) annual grant for stationery and books (±R7,000 -R8,000); d) covers costs involved in teaching internships; and e) ±R 600 monthly stipend to cover basic living expenses. The Brazilian Prouni provides bursaries for any public sector teacher, from primary or secondary education, who has not a higher education degree. The range of undergraduate courses supported by program is limited to education, social sciences, history, biology, geography, mathematics, and physics. However, there is no requirement for a maximum per capita household income for the student to be eligible for PROUNI´s teacher bursaries.
Once again, South Africa's program for teaching focuses not only in access to high education level, but to promote allowances so that students can successfully complete their undergraduate courses, such a mechanism is absent in Brazil´s Prouni. The number of bursaries awarded in each program is very different. Between 2007 and 2017, the Funza Lushaka bursary programme awarded 120,511 bursaries, whereas Prouni for teaching awarded only 12,225 bursaries (from 2004 to 2014). Considering the differences between the number of enrollments in higher education system in Brazil and South Africa (see Graph 2), the relative importance of Funza Lushaka bursary programme is much higher than Prouni.

International Rankings
Beside the expansion and inclusive policies implemented in both countries, many forms of assessing university quality became an important subject especially as international rankings became a popular method for evaluating these institutions. Using diverse methodologies and criteria, there is still a lot of argument or debate about why and how to consider the rankings both for managing the institutions or choosing to enroll in one of them (Clarke, 2002). Most important, the debate highlighted the race in course to constitute "World Class Universities" in each country and this is seen as an indicator of the development of a market for international higher education (Martins, 2015). The Times Higher Education ranking of universities in emerging countries for 2018 allows for some initial comparisons on this topic (Table 1). The rankings eventually raise a lot of debates, especially as they favor Englishlanguage domination and institutions over others. Anyway, they can be read as indicators of the level of success of each country in creating institutions that could be defined as being "world class universities". Therefore, considering this framework of great enrollment increase together with several policies to open up access and facilitate the permanence of students in the higher education system one must ask to what extent Brazil and South Africa have managed to improve both international quality rankings positions and indicators of equality of educational opportunities?

Theoretical Approach
Societies build models for democratization of their educational systems, in ways that result from configurations of social forces, which in turn produce historical trajectories. In this article, we shall use parameters established by Coleman (1968) indicating that equality of educational opportunities means ensuring every citizen, regardless of his/her class, race or gender, can obtain entry into effectively accessible institutions, qualified learning and training in a common curricular framework. Therefore, equality of educational opportunities means that the school system -or, in our case, the higher education system -precludes social determinism, ensuring that all young people are capable of learning and that their performance reflects their efforts and intelligence and not their social origins. We consider a higher education system to be more democratic when it guarantees greater opportunities to the population -tending to equality. The conceptualization of phases of higher education systems elaborated by Martin Trow (2007) has also been used in studies of Latin America and Africa (Hornsby & Osman, 2014). A system's evolution is characterized by the proportion of students they would be able to enroll: elite systems are those where up to 15% of the correct age bracket are enrolled, and in a universal systems this figure is over 50%. We define a mass higher education system as one that enrolls from 16% to 50% of individuals in the age bracket appropriate for their entry into tertiary education.
The massification of a HE system can improve access of some groups such as cultural or ethnic minorities and women. However, massification is different to democratization. Not all social categories benefit in the same way from massification. Even with more students entering universities, the democratization of access also depends on the general structure of the education system. The higher education systems tend to develop a type of academic meritocracy that would build a hierarchy of skills. This is not only a social hierarchy itself, and strongly contributes to the reproduction of wider social hierarchies (Dubet, 2015, p. 258).
The theories of social closure developed by Weber (2013) and Parkin (1979) were systematically used to analyze various strategies for achieving and maintaining power and social status, especially by those who dominate the educational system (Karabel, 1984(Karabel, /2005. The theoretical and methodological improvement and refinement of this concept (Lareau, 2011;Van Zanten, 2015) identifies the complexity and subtlety of the social processes, which involve power struggles, in an area, which is supposed to be meritocratic and impartial. Precisely due to the assumption that the university would be the expression of modern forms of legal rational domination, the sustained presence of those linked to traditional forms of authority in these institutions leads us to revive the Weberian concept of patrimonialism (Lachmann, 2011;Charrad & Adams, 2011;Campante, 2003). Similar to the institutional matrix of scientific rationality in contemporary societies (Parsons, 1967), the university, and tertiary education as a whole, represent the apex of democratization given their use of merit-based ranking criteria. However, due to the persistence of, and increase of inequalities when access to higher education is expanding (Alon, 2009;Dubet et al., 2010) idealized approaches lack credibility. Weberian concepts related to forms of domination seem to offer more adequate explanations for the struggles and disputes over the meaning and value of higher education (Weber, 2013;Szelenyi, 2016). Through these concepts, it is possible to understand why teaching or scientific research are not just technical matters. They also constitute a more or less legitimate form of the exercise of power. From this perspective, academic activities produce forms of social hierarchies of skills. A tradition of sociological research perceives the higher education system as a set of formal institutions that participate in these struggles and are subject to pressures and social expectations, while not denying that they have a certain degree of autonomy (Coté & Furlong, 2016;Gripp & Barbosa, 2014). Furthermore, the specific functioning of these institutions is a key factor in increasing or reducing inequality in higher education.
The dispute over whether to include more students or to create world-class institutions will be analyzed in the perspective outlined above. Winners in these disputes would establish legal and institutional parameters that allow for different levels of system openness and/or high-quality institutions. While they are not mutually exclusive, the option or the preference for one of these models has impacts on the functioning and results of the higher education system.
As in other countries, and especially in the United States, Brazilian sociological research tried to analyze the key institutional features that organize the country's higher education system. The pioneering work undertaken at NUPP (Center for Research in Public Policy) at the University of São Paulo covered various topics such as scientific development, university autonomy and the academic profession 5 . These studies are the initial base of research into the Brazilian higher education system. Later on, other centers were set up and the Ministry of Education, through the INEP (National Institute of Educational Studies and Research Anisio Teixeira), currently centralizes the collection and systematization of data and the evaluation of higher education. Highlighting the strength of the so-called academic drift in the Brazilian tertiary education system was one of the main results of NUPPs' research. Schwartzman (2011, p. 15) uses the term "academic drift" to describe the attempts of educational institutions to increase their status by imitating the most prestigious organizational models and areas of knowledge, this strategy reduces diversity. Its effects extend down to the lower levels of education with important consequences for the structure of inequality in Brazil. There are indications that more than being a simple bias, academicism is the dominant feature of the Brazilian higher educational system (Prates & Barbosa, 2015). More recently this academicism came to be reinforced by national institutions being encouraged to seek to acquire the status of "world class universities" (Martins, 2015). Perhaps this academic bias is simply a result of the Brazilian law that defines one single model of tertiary education, which is based on the large research universities.
Forms of social and political organization in Brazil tend to be traditionalist (Faoro, 1998;Schwartzman, 1988). Even during strong periods of economic growth, which marked Brazilian modernization, which included significant urbanization and industrialization. Our university system was created late and, with rare exceptions, without having modern scientific knowledge as its major reference (Barbosa, 2012). Little value is attached to education as the basis of occupational, professional, and social achievement (Barbosa, 1996;Almeida, 2007). Although there is evidence of a "civilizing" impact of the passage through the university Almeida (2007) shows that students emerge less racist, sexist, and discriminatory than their peers do. Lack of education is often mobilized to justify a person's failure or poor realizations in the labor market. However, when people legitimize successful trajectories they rarely mention their educational achievements! In South Africa, a new constitution was established in 1994 in the process of democratization subsequent to the overthrow of apartheid and its unique structure of social, economic and political inequalities. The new government focused on higher education, producing an array of institutional changes, which sought to enlarge and diversify access. Some studies and proposals for policies and institutional actions came from research centers at some universities (Johannesburg, for instance). However, the full organization of the educational system came after "the development of policy and legislative frameworks, [in] the second period of government (1999)(2000)(2001)(2002)(2003)(2004) [that] saw an elaboration of policies which further enabled the government to get a grip on the levers of power in order to steer the system to not only to be able to deal and respond to global challenges, but also to be locally responsive and relevant" (Sehoole, 2011, p. 977). New legislation, always based on the values and principles of the emerging democracy, focused at increasing participation, improving responsiveness to social and economic needs, and seeking cooperative governance. (Sehoole, 2011, p. 978). The whole higher education system was restructured, public and private institutions were centrally regulated, and accreditation programs were developed. The previous fragmentation was overcome: the 1997 White Paper established a higher education branch at the Department of Education (that eventually gave birth to the Department of Higher Education and Training (DHET) and the Council on Higher Education (CHE)). Both institutions, using data from STATS SA (The statistics institute for South Africa, created in 1976), provide the best information on policies and results of higher education in the country.
The hypothesis that guides this study, based on the concepts proposed above, is that traditional and colonial forms of authority are a constituent part of the higher education system. This article tries to show that these patterns or institutional models seem to work as barriers to the effective democratization of educational opportunities. These barriers could act as obstacles even in a context of public policies directed towards social inclusion and to making very significant increases in enrollment in higher education.
To examine this hypothesis previous studies developed by the Brazilian research group (LAPES -Laboratory for Research on Higher Education) and by other researchers were used. Data from Brazilian Census of Higher Education 2013 and 2014 and data derived from ENADE 2014 (National Student Performance Exam) are all produced by INEP. This quantitative data permits some generalizations to be made. However, some case studies are also used in this article. For South Africa, we use data from STATS SA. The institution counts two million higher education students. Nevertheless, none of South African statistics includes Vocational Education and Training (VET), Medicine and courses in the area of Health, nor do they have information on private institutions in the data and tables on enrollments or graduation rates for Higher Education. Because of that, there are significant differences in the numbers that will be presented in the article. We chose to use only the available data on the public universities (traditional, comprehensive, or technological) as they were presented by The Council on Higher Education, which is considered the official source (along with STATS SA) in the country.

Socioeconomic and Historical Context
In Brazil and South Africa, as in many countries, there are quite substantial individual economic returns to schooling, which would be a normal outcome, considering the progress made in terms of modernization of social relations in both countries. However, the difference between the individual returns for different levels of education is changing and, ultimately, is making room for questions about the very legitimacy of education as a criterion for the distribution of socioeconomic positions. As shown in Menezes' studies (e.g. Menezes & Pecora, 2014), the balance in the returns for different levels of schooling is changing with a reduction in returns at the higher levels. This distinction also allows us to understand why access to higher education has become an important object of social demands.
The returns to education in South Africa are significant, being very unequally distributed among different racial groups. Higher education works as an important factor in reducing unemployment from about 40% among those with only primary school to about 6% for graduates (Mapadimeng, 2017).
The expansion of higher education in Brazil began very slowly during the military governments but gained momentum from the year 1990. As the data from INEP (2014) shows the growth rate in 1980-2000, was 96% and for the period 2000-2014 was 190%. Counting nearly 8 million enrollees in 2014 (see Graph 1), the gross rate (or participation) of Brazilian higher education coverage reaches a mere 34.2%. This is already a problematic indicator: considering only those aged between 18 and 24 (the ideal age cohort to attend higher education), the 2010 demographic census shows a participation (or net enrolment rate) of 13.94%. In other words, less than 14% of young people in this age group were enrolled in some kind of post-secondary education. In South Africa, the numbers appear somewhat similar to those in Brazil: just 12.1% of the population aged 25-64 had a higher education degree in 2016, according to STATS SA. According to the same institution, in 2016 the participation rate reached 18% in South Africa. In Trow's classification, both countries are midway through a transformation from an elitist to a mass system of higher education. Even considering their recent expansion.
Brazilian growth in enrollment was more intense in private institutions. While the public sector grew by 80.5%  and 120. 7% (2000-2014), the rates for the private sector were 104.1% and 224.6% over the same periods. Of the two million post-secondary education students reported during 2016 in South Africa (including VET ones), 78.3% were enrolled at a public institution, whereas 21.7% attended private institutions. According to CHE, an average of 110-120 higher education private institutions operated in South Africa between 2002 and 2016. Considering the most recent data, there are 2,364 higher education institutions in Brazil, of which 87% are private ones. Private colleges cater to 75% of all undergraduates, most of whom attend for-profit institutions (55, 33% of all students) (Higher Education Census, 2015).
The next graph shows enrollment distribution per type of academic degree or track. This data makes clear the utmost preference for the bachelor's degree, which are responsible for approximately 67% of enrollments over the above-mentioned period. In the same period, enrollment in degrees to prepare for a teaching career seems to have undergone a slight decline, despite public policies to the contrary. At the same time, tertiary-level technological courses have had stronger growth, having multiplied their share in total enrollment six-fold. The courses that offer technological degrees and teachers' degrees were legalized only in 1996, with the new LDB (Law of Directives and Bases for Brazilian Education).
The similarly stratified distribution of degrees in South Africa is presented in the next figure. Diplomas (Occupational certificate level 6) and certificates (Occupational

Source: CHE VitalStats -Public Higher Education 2016.
In Brazil, the enrollment ratio in the STEM (Sciences, Technology and Mathematics) varies from 17% to 23% over the period, and in the area of Education it reduced from 21% to 19%. In South Africa, the figures are similar for Education but in 2016 STEM reaches 30.27%, almost a third of all students.
There are 32 bachelors' degrees offered in South Africa. Unfortunately, we were unable to find statistical information available on the topic on CHE or STATS SA sites because their data refers only to public sector higher education. Also, differently from Brazil, enrollment in the health sciences area is not included. This introduces a problem for comparing the proportions of students in each area.
The common trait in the two cases is the high participation of business, education, and humanities areas. In the case of Brazil, even including enrollments in the health area, business is the area chosen by almost half of the students.

The Brazilian System of Higher Education
The Brazilian system of higher education was developed late, even when compared to other Latin American countries or other BRICS countries. The first Brazilian university was created in 1920, the University of Brazil. The federal government joined three isolated colleges under the title of university in order to bestow the title of doctor honoris causa to the King of Belgium during his visit. It remained the model for many states/provinces from the 1920s to the 1950s. The University of São Paulo, created in 1934, was the first attempt to have an institution conceived to work as a higher education structure and not an aggregate of smaller colleges under a single administration. The BRICS countries show significant similarities in their economic development process: in all cases, we see the prevailing acute shortage of resources coupled with a close relationship between the higher education system and the political forces that control the State in each of these societies (Schwartzman, 2015). According to the same study, the five BRICS countries passed through a strong institutional diversification, implemented affirmative action policies of several kinds and their internationalization policies have not been very successful. The increase in enrollment occurred mainly in social sciences, humanities, and education. Some features of the Brazilian system are noteworthy: by a constitutional requirement, public institutions do not charge for enrollment, they are totally free. This is a unique trait to Brazil, when compared to the other BRICS countries. Public institutions, especially the state-funded universities in São Paulo, tend to be evaluated more positively, both in internal classifications made by government agencies and in international rankings. However, even with a system of higher education that has bureaucratic, technical and social diversity, Brazilian law imposes a single model to be followed by all institutions. Through legislation, research universities with many graduate courses, a small college or a training center for industry should work under the classical model of universities proposed by Wilhelm von Humboldt (Schwartzman, 2014, p. 23). This legal demand obviously has impacts on the possible paths towards democratization of the higher education system. Each institution should be prepared to function as a "world class university". One can understand this definition of a single legal model for the whole system of higher education as a handicap. Although established by bureaucratic and legal rational parameters, this traditional model of university reduces the range of socially valued training options. The diversity and experimentation that has characterized tertiary education elsewhere, for example in the United States, Germany or South Africa, is excluded from the alternatives offered to Brazilians. The ideal university is prescribed as an advanced research institution, with high quality post-graduate programs. Undergraduate courses follow the model of bachelor formation; they invest strongly in theoretical knowledge and provide almost no practical training. This appears to be a restrictive model when one considers the demands of the labor market, the needs for teacher training and the expectations of young people seeking higher education. This legally defined model of a university highlights the strong academic bias that runs throughout the education system. So, instead of increasing qualification opportunities and certification alternatives, the legally defined higher education model, raises the bachelor's diploma to the position of "general equivalent" (Thévenot, 1983), the parameter for measuring of all qualities. Therefore, this higher education system works in a specific legal framework that defines its bureaucratic rules and its institutional excellence models. There are three defined levels in the system: the lowest one is the "isolated college", a small institution with few courses. The next level is the "university center": "institutions of higher education, covering one or more areas of knowledge, characterized by the excellence of teaching offered, proven by the qualification of its faculty and by the academic working conditions offered to the school community. These "accredited university centers" have the autonomy to create, organize and extinguish, at their headquarters, courses and higher education programs" (Brazil, MEC, Decree nº 5.773/06). Finally, the term "university" defines the highest level and organizes the hierarchy of higher education institutions. "Universities are characterized by the indissociability of teaching, research and extension activities. They are multidisciplinary institutions for the training of professional staff at a high level, research, extension and the mastery and cultivation of human knowledge, which are characterized by: (I) -institutionalized intellectual production through the systematic study of the most relevant themes and problems, from the scientific and cultural viewpoints be they regional or national; (II) -one-third of the teaching staff, at least, with an academic master's or doctorate degree; and (III) -one third of the faculty on a full-time basis" (idem ibidem).
This conception of what a university should be is associated with the Humboldtian model whose core idea is a holistic combination of research and teaching. The model integrates the arts and sciences with research to achieve both comprehensive general learning and cultural knowledge. According to Anderson (2010), this kind of model shaped the research universities in the United States and all around the world. "The Humboldtian university can be seen as the characteristic form of the university idea until the growth of mass higher education in the late twentieth century" (Anderson, 2010). The adherence to this type of model is very problematic. The Humboldt model designs elite universities and, historically, could be opposed to the conception of mass higher education. In this way, the Brazilian higher education system suffers from a paradox: it has the ambition to reach large segments of the population offering a model theoretically aimed at teaching the elite.
This model of accreditation and hierarchy makes any institution, even those smaller or characteristically more vocationally-oriented, work as a "university under construction". All of them aim of being eventually recognized as a university which conducts both research and academic formation. Many of them become mere simulacra of a university. As is easy to imagine, the social distribution of the population throughout these different types of institutions is not random.
This same model is required for public and private institutions, for courses that offer bachelor's degrees, or teaching degrees, or even technology courses! However, it is not difficult to find, within this single legal model, several types of segmentation that end up producing enormous differences in institutional functioning and the quality of the services offered.

South African System of Higher Education
The University of the Cape of the Good Hope, created in 1873, was the first in South Africa. Established in the southern part of the country, it eventually became the University of South Africa (UNISA). With the universities of Stellenbosch and Cape Town, UNISA could be linked back to the settlement by white colonialists in the 17 th century. The higher education system developed during most of the 20th century was marked by significant racial differences. Around the 1960's only 5,000 of the 62,000 university students in South Africa were non-white. According to Sehoole (2011), the racial disparity was followed by a gradual "racial opening up" and the figures changed. By 1980 there were 150,000 whites and 120,000 non-whites enrolled. In the first decade of this century, the majority (67%) of students in the 21 public universities are non-white. The higher education system had a binary structure of traditional universities and technikons (institutions for technical higher education). The latter evolved from the old technical education institutes, favored mainly by the mining industry. To address the needs of this industry, training centers were established as apprenticeships. After some decades, with the inclusion of more abstract items in the curriculum, the centers became colleges. The Higher Education Act of 1923 declared some of them as higher education institutions. In 1970, the Ministry of Education recognized these technical institutions as constituent part of the higher education system, with equal status to the traditional universities. Only in 1993, the Technikons were authorized to award degrees as any other university. In 1994, the new democratic government "inherited a well-established higher education system made up of 21 universities and 15 technikons" (Sehoole, 2011, p. 974). All of them public institutions. In 2018, the private sector comprises roughly 10% of enrollments.

A Segmented System: Public or Private Institutions in Brazil, Black and White Universities in South Africa
The Brazilian private and public sectors are deeply differentiated (Sampaio, 2014), even in their institutional characteristics. Thus, among public institutions, there are federal, state and municipal entities, defined according to the legal responsibility for their management and financing. It is important to remember that in Brazil, public institutions have free tuition, with all costs covered by public funds. On the other hand, private institutions are distinguished by being either "for profit" or "not for profit". Among the latter, there are the so-called Community Institutions -often associated with religious groups -which sometimes attain very high levels of quality (Neves et al, 2008).
Private institutions receive 74% of students' enrolment in Brazilian HE. Some public policies and institutional actions were developed to assure the ability of enrolled poor students to remain studying in these universities. The costs for the best institutions and enrollment in courses leading to the most prestigious careers are very high: the average monthly fee for a medicine course in a private university is approximately R$ 7,000 (during 8 months of the year, approximately US$ 1,800 or 7 Brazilian minimum monthly wages) -an amount beyond the reach of most Brazilians who are also unable to succeed in the very competitive entrance exams for the public universities. medical course fees start at R$4,500 per month and reach up to around R$13,500! And, this is just the cost of enrollment: it does not include the acquisition of books or research materials, transportation, and lodging. There are, for sure, less expensive courses, mainly in the area of Humanities (e.g. Philosophy, Sociology, Geography, Psychology etc.) or Education (which includes Pedagogy and all teaching degrees). The monthly fee in a Pedagogy course could be as low as R$270 when conducted by distance education. Anyway, for many fees are a big expense in a country where annual per capita income is US$15,700 (in 2015).
Policies for financing private higher education have been developed over recent years and are now very relevant, two major programs were established: PROUNI is a policy that offers full and partial scholarships for poor and/or black students through tax exemptions for private institutions; FIES is a student loan fund with reduced interest rates that must be repaid in the long term. The data demonstrate the importance of these mechanisms. In the period from 2010 to 2014, enrollments in the private sector grew by 17% (Table 2), while students benefiting from FIES increased by 750%. This explosive increase is associated to political strategies and suffered a significant setback in 2015 and 2016 due to cuts in government funding. But it remains one of the main instruments of access to private higher education in the country.
There are also social welfare policies for poor students even in the tuition free public universities. Students entering public universities through quotas or other affirmative action policies receive various types of economic aid. They are mostly fruit of policies developed within these institutions that aim at ensuring -with greater or lesser success -the permanence of these students.
In addition, the federal government's REUNI program aimed to expand the public system of higher education. According to data from the Ministry of Education, 14 new universities and 100 new campuses were created, thereby expanding the federal network. In this context, a significant expansion of enrollment in federal institutions occurred. In 2007, when REUNI was created, these institutions had 12.61% of enrollments in higher education. This proportion rose to 16.71% in 2014. In addition, the creation of evening courses, especially in the area of education, has increased the participation of less affluent and working students.
In Brazilian society, several studies that indicate an aversion to the private sector and an appreciation of the public sector (Almeida, 2007). Coupled with the fact that the best Brazilian universities are public, this evaluation flows on to all institutions in the sector, regardless of their actual quality, and likewise diminishes the expectations held of their private sector counterparts. Yet, as shown above, the private sector is by far the most important for enrollments and their offer of courses is significantly larger: 10,240 in the public sector in 2014 versus 21,842 in the private sector. The public and private sectors have been distinguished along several dimensions: in this section, we discuss some social characteristics that distinguish the students who attend the institutions in each of the sectors. First, the average age of students enrolled in higher education differs significantly. For public sector students, the average age (Brazilian Census of Higher Education, 2014) is 25.97 years (standard deviation 7.764), while in private institutions the age rises by nearly two years to 27.83 years (with an even higher standard deviation: 8.391). For 2013, the year that ENADE examined agronomy and social work students, in addition to all specialties in the health area, the age gap at graduation was three years. Going to HE at a later age is usually associated with a more modest social origin, or with entering courses of a more vocational nature (Connor et al., 2001, p. 106;Prates & Collares, 2015).
In Brazilian society, the preferences for the type of degree are fairly marked from the social point of view: Teaching degrees or technological courses have been "preferred" by the students from more modest social backgrounds. The data above confirm the studies that indicated this trend with a clear increase in demand for the less prestigious courses and degrees for students from less affluent social groups. As for the students of the educated middle classes, the choice falls mainly on the very prestigious courses and with a very high number of students competing for each place offered in the public institutions. These preferences for bachelor's degree are understandable. Having a bachelor diploma entitles a person to work in the public sector and to receive a premium wage.
In South Africa, the hopes of democratization of HE system subsequent to substantial changes in the legal and institutional framework did not materialize. Contrary to expectations, students left historically black universities (HBU) for the historically white ones (HWU) where they could find better infrastructure and teaching or more chances of funding for their studies. HWU also offered better opportunities in the job market. It was truly a mass migration phenomenon. The best institutions in the HE system sized the opportunities to better their positions, including investing in distance education. The Higher Education Act (1997), permitted private institutions, and they experienced huge and almost unregulated growth. There were no registration nor quality requirements. The crisis that emerged subsequently induced the consolidation of a policy and regulatory framework aimed at the constitution of a single coordinated system. (Sehoole, 2011) In the case of Brazil, since the 19 th century, the private sector has been strongly regulated by federal government, which has a great deal of control over Brazilian higher education system, including the public institutions.
Emerging from a more mature and organized governmental structure (in the period 1999-2004), policies were designed by South African professionals with the participation of the ministries responsible for the Economy and Education. The Higher Education Act was amended in order to legally reorder the multifaceted and unorganized set of higher education institutions. The new single system of higher education implied a consolidation of the public sector from 36 to 21 institutions, the legal regulation of private providers that made their registration at the Department obligatory. One important measure was that the South African Qualifications Authority must assess all qualifications standards. Another dimension was the accreditation of programs, especially post-graduate ones. The assessment of MBA programs in 2003 indicated many problems, mainly among international providers and Technikons. This kind of problem, as noted by Sehoole (2011, p. 990), made South Africa a pioneer country in developing a comprehensive framework capable of dealing with the challenges of cross-border or transnational systems of education.

Discussion
The expansion of Brazilian higher education system effectively allowed some less affluent social groups access to tertiary level education. Given the previous low coverage levels, this was certainly an important step towards the system's greater democratization. The same could be said about South Africa post-apartheid, when African, Colored, and Indian students got more opportunities to enter the best universities.
However, considering the net coverage rate, around 14% and 18% respectively, it can be said that this expansion is still quite restrictive and less than inclusive, or at least not sufficiently inclusive to allow Brazil and South Africa to be ranked as relatively democratic HE systems. Statistically, enrollment remains, in both countries, very close to Throw's definition (1970) of an elite university system. The problem that arises is similar to Alon's (2009): how to explain that in the context of expansion, in spite of public policies focused on the democratization of access and permanence, HE remains relatively closed, particularly in courses that give access to more prestigious and well-paid careers. Both in Brazil and South Africa, it seems that the theories of maximum maintained or effectively maintained inequality (MMI and EMI respectively) fit, as has been shown by many authors for the United States and Israel. Diversification and expansion did not keep pace with democratization. Even worse, it seems that diversification is a way of diverting more disadvantaged students from the more privileged professions or careers (Shavit & Blossfeld, 1993).
For the Brazilian case, the data presented below indicate both some advances associated with the higher education system and some bottlenecks in opening educational opportunities for Afro-descendants (previously called "blacks") and for women. The proportion of Afro-descendants remains very low in the most prestigious careers and with better social status and economic returns. At the same time, women are now the majority of higher education students but enroll in courses that lead them into the lower paid occupations. The data on family income in Table 3 seem to indicate that higher paid professionals come from affluent families and those who earn less come from poorer families.  It was not possible to find similar data for South Africa. However, the numbers on completion according to the race, presented in Table 4, show that Africans, Colored, and Indian students have lower rates of graduation than their White peers. Even 22 years after the end of apartheid, it seems that the non-white students still face some difficulties in their higher education trajectories.
Our analysis, following the tradition of sociological research, understands that the mode of operation of higher education institutions stands out as one of the key factors in the mechanisms and social disputes that increase or reduce inequalities. Focusing on the basic distinction between public and private sector, we try to show some of the effects of this segmentation and, maybe, of academic drift in Brazil. More than a bias, academicism seems to be a dominant feature of the higher education system either because of the legal definition that imposes a single model, or because of the rules and practices of diverse social agents. This is shown not only in the ways of teaching and in the definition of the subject contents, but also in the adoption of a perspective that devalues nearly all practical, technical, and even scientific knowledge.
Probably the strongest indication of the academic bias is the generalized preference for bachelor's degrees.
The private sector of higher education has more students and courses and, in a slightly different manner to the public sector, it invests heavily in technological courses. This may indicate a lesser impact of academic bias in this sector. Data on graduates shows that, overall, the higher education system remains a very elitist area, but that there are slight openings in the private sector.
In South Africa, after the end of apartheid, many Africans moved to the historical white (and better) universities, trying to get more privileged positions in the job market (Sehoole, 2011). This made it more difficult to implement policies to improve the quality of the historically black universities, and even more so to develop world-class universities among them. The best universities in South Africa are still those traditional institutions, which are historically white. The South African system of higher education is more accessible to those groups of disadvantaged students, but as in the case of Brazil, these students seem to be diverted into the less privileged careers, such as technological courses or in the area of Education. Women are the majority of students, except in the STEM area. However, the data does not permit us to see if their participation is also translated into better positions in society and job market.
Both Brazil and South Africa have invested in improving the equality of opportunities and the quality of higher education. The Times Higher Education rank for emergent countries showed that South Africa can be considered more successful than most others in its policies to create world-class universities. Even so, in both countries the quest for excellence seems to be much more related to the emphasis given by administrators to bureaucratic procedures than to academic investments (Martins, 2015;Govender, 2018).
Although the data is incomplete, there are some indications that, measured by the rates of completion presented by CHE, South Africa is ahead of Brazil in opening up its higher education system. Again, after institutional and curriculum transformation, policies for social and economic priorities at the higher education level, social inequality and structural contradictions are reproduced within existing power relations with the strong contribution of the higher education system.
The two countries appear to be very comparable with respect to the difficulties faced in opening and democratizing their higher education systems. Gender and racial inequalities still permeate trajectories and achievements in both systems. In Brazil and South Africa, the investments in scientific formation (and in the STEM area), that would improve participation in knowledge society, are precarious and most students are enrolled in business and the humanities.
Even considering the problems with the statistical sources, South Africa seems to be a little ahead of Brazil both in opening up and improving the quality of its HE system. One hypothesis to explain the small differences between two countries which otherwise appear to be so similar would be the academic bias, so strong in the case of Brazil, having a much less significant role in South Africa. The development of a more modern HE system, one that is democratic and scientific, becomes difficult when the existing system is very strongly influenced by traditional and patrimonial values. Science and modernity do not go well together with tradition and patrimonialism.