Why is Brazil so unequal Looking at the bare numbers in the preceding subchapter the question arises what could possibly be the reasons for Brazil's enormous inequality. Income in- equality exists because people differ in all kind of aspects that are relevant to the income they achieve. What usually contribute most to income inequality are differences in huluan capital, i.e. education Those differences in humau capital will result in different returns on the labour market. The more unequal the distribution of human capital and the higher the return to educa- tion, i e. a skill premium, the more unequal will also the income be distributed. Skill premiums are driven by supply and demand on the labour tnarket and on productivity a worker can achieve (Weil, 2012). However, income inequal ity can also have different sources: there can be geographical diferences, eg. Situation in Brazil When looking at possible channels that influenced Brazil the most credit market imperfections should probably be the one chamel emphasized the most, especially in the pre-democracy era before 1985. During the period of military
dictatorship during 1964 until 1985 it is reasonable to presume some kind of favouritism towards a specific group that was leaning towards the rulers of this time. This alone might already be an inequality fact, but further it's realistic to assume this in-favour group had better access to the credit-market, e.g. loans. After 1985 as institutions grew more reliable and democratie right came in to place, the number of loans in general probably went up. Whether it changed the situation for the poor with little or no assets to get their loans for investing in mostly their human capital or simple machines for agricultural activities is difficult to tell. However, with the emergence of micro-credits in the last couple of decades and the subsidizing of such micro-credits through the government since 2005 to bring interests rates down to 8%8 it can be assumed that today more poor people can have access to credit-market institutions than in the past. Therefore, I would judge the channel of credit-market imperfections as important for Brazil but with declining importance in the three decades last and possible in the future as well.
Applying the logic from Chapter 3.2, considering the high income inequality and the democratic situation in today's Brazil, one might guess that huge re- distribution programs were in place to partly balance out the coun income. Immervoll et al. (2005), though, provide facts that contradict that theory. At the time of their study, the tax-system in Brazil had a very little equalising effect on income inequality, mainly due to its regressive pension program. This signalises that the channel is of lesser importance although the tax revenue as percentage of GDP is at 39.9 9 which is a high number in international compari- son. This can provoke economic distortions and inefficiencies but high inequality is unlikely the sole trigger for the high tax revenue number.
Political unrest in Brazil today is not an issue that could have huge effects as a channel, even during and at the end of the period of military dictatorship there is no evidence of civil war-like circumstances with highly explosive politi- cal riots and excesses (Skidmore, 1990). Moreover, with the economic advances the country experienced in the last decade, political unrest seems even more un- likely. Although inequality has stagnated for a long period, wages for low-skilled workers, pensions and entitlements for the poor have still risen in absolute terms. This should also have an effect on poverty-based crime rates as the incentive
for doing something illegal, e.g. for stealing, has decreased because of better outside opportunities, i e. people who are considering illegal activities have now more to lose than before. Sachsida et al. (2010) have explored the connection between crime and inequality in Brazil and found that high increases inequality criminal behaviour but found no evidence on a relationship between inequality and strongly violent crimes. But again, it's very difficult to distinguish whether these criminal activities are induced by inequality or just absolute poverty lev els. On one hand, it's hard to believe that when in a theoretical approach, all people would have their income doubled over night (but prices on necessities remain constant), thus leaving inequality exactly the same, that no one changes his decision to enter in to illegal activities. On the other hand, several stud- ies like Ravallion (2001) have showed that inequality and poverty numbers are heavily correlated.
To my knowledge, no data on income-group specific saving rates exist for the case of Brazil. But results from international research regarding the topic suggest that tend to rise significantly with income. Dynan et al. (2004) collected data for households in the United States where at least one member was aged 30-59. The results showed that the lowest income quintile saves only 9% of their income, while the rate of the highest income quintile is at 24.4%. It is plausible to assume a similar outcome in Brazil, especially because the poor in Brazil are in absolute terms worse off than in the United States and have therefore even less room to generate savings. After looking at the results from Dynau et al. (2004) and the suspicion that results could very well be more extreme for Brazil, the channel of different saving rates for different income groups cau play a big role in the effect of income inequality uri economic growth. However, with the trend of globalization and with economies getting tuore open to foreigu investment. a country's saving rate has less effe iu determining the country's amount of iuvestineut which will help produce corrobuie growth.
Whit a rate of population living under the poverty line at 26%10 there exists considerable high number of people who can barely survive on their eurrent economic situation. Such extreme conditions can lead a poverty trap (Baner to jee and Duflo, 2011) caused by insufficient nutrition or health conditions. The person who expierences such a poverty trap finds himself in a vicious cycle from which he is not able to escape, e.g. not being healthy leads to an inability to work, which prevents him to get the income needed to pay for adequate treat- ment of his health condition. As high inequality can be a cause of poverty or is at least strongly correlated to it, this channel could prove to be very impor tant. A person who experiences the situation of being in a poverty trap will not contribute to any economic growth, so the more people that are able to escape those conditions, the more people will particape in economic activities, which will lead to higher economic growth.
From 1960 until 2000 Brazil experienced a sharp decline in fertility overall ter et al., 2010), but the development was not evenly distributed among different income groups (Muniz, 2009) Fertility first declined for the upper income groups, thus increasing the fertility differential between the rich and the poor. But in the last decade covered by the research of Potter et al. (2010) fertility has also started to decline for lower income groups which, together with stagnating number of fertility for upper income groups, has decreased the fertil ity differential again. Therefore, one can conelude that inequality has affected economic growth through the channel of fertility differential among the poor and the rich more in the past than it does now. Still, it is also today a factor that cannot be underestimated After analysing the importance of the different possible channels, it is hard to conclude which effects will have a dominating influence in the ease of Brazil. Alone from this qualitative analysis it is impossible to guess the net effect t inequality has on growth. Therefore, in Section 5, the paper will assess the issue empirically and try to call the effect the Gini coefficient has on growth in the subsequent period.