Perspectives on Democracy and Inequality:
Government, Health and Education.
Gijs Schot
January 31, 2017
Abstract
What is the effect of democracy on inequality? This debated research question and topic is characterised by unambiguousness and complexity. Different theories give reason to believe the relationship between democ-racy and inequality should be positive and this is supported by many different studies. Causality however seems harder to proof and some ar-gue democracy not a causal factor at all. Answers that explain the true relationships at play will most likely come from creative research design. Rigidity in the way research is executed seems the biggest hold-back in finding solid conclusions in the paradigm as of this moment.
Statement of Originality
This document is written by Student Gijs Schot who declares to take full re-sponsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the super-vision of completion of the work, not for the contents.
Contents
1 Introduction 4
2 Democracy and Inequality 6
2.1 Democracy and Government . . . 8 2.2 Democracy and Health outcomes . . . 11 2.3 Democracy and Education . . . 12
3 Conclusion 16
1
Introduction
Inequality matters: in 2010 almost half of the world population lived on less than $2 a day. About three million babies die each year, which could be easily prevented by cheap measures. Inequality is not only relevant in poorer na-tions, also in the developed world inequality is on the rise. Many economists have therefore made contributions in providing perspectives on the relevance of regime type regarding their relative influence on inequality in society. The effect democracy has on inequality is a debated topic due to its unambiguousness and complexity. The divided views within the literature call for an examination of the status quo in this paradigm. Recent meta-studies with focus on the content of inequality and democracy are nonexistent as far as my knowledge goes. With a systematic approach I will map the patterns, empiric traditions and knowledge of selected scientific articles in order to provide an overview that can bring new insights to the topic of democracy and inequality.
Many economists support theories that state that democracy is decreasing inequality. The median voter theorem being one of the most recurrent theories. The median voter theorem states that in a voting system that follows a majority voting rule, which is true for most democracies, the outcome that is preferred by the median voter will be selected. The theory explains why political parties at the end of the political spectrum are not likely to be elected. The theory also predicts that if the income distribution in a democracy is as follows: the mean income is higher than the median income, you can expect the median voter voting pro redistribution. In that way inequality will be decreased.
Holcombe 1989 finds strong empirical support for the median voter theo-rem. Empirical support for the fall of income inequality in democracies however remains much more disputed. Multiple studies have confirmed a positive corre-lation or even claimed causal recorre-lations from democracy to equality. Many other studies however attenuate the robustness and validity of these correlations and causalities. A meta-analysis is a powerful tool for reviewing an already well and broad investigated research field that does not reach a clear consensus on the implications of the available data. Therefore the main question I try to answer in this paper is: what general conclusion can be made about the effects of democracy on inequality?
In the process of gathering the literature, I noticed that most studies laid emphasis on one or more of the sub effects: government, health and education and their relation to inequality and democracy. Thus leading to the categoriza-tion in these three subjects. Within this field, inequality can be measured and characterised in different ways. Welfare programs, education spending, health conditions within society and plain income inequality in the form of a Gini-coefficient for example all address inequality in their own manner. By choosing the three subject structure I hope to give a clear outlook on the topic of in-equality. An important aim is to give guidance in this field of research where incoherent and contradicting conclusions make it sometimes difficult to figure out which relations are supported and which are not.
Throughout the studies under review democracy is generally measured in two different ways. The first method is using a dichotomous measure, a country regime is either a democracy or a dictatorship. In some studies the dichotomous measure becomes a trichotomous measure by adding the possibility for hybrid regimes. The second method of measuring democracy is using the Polity index.
This index scales countries regimes from minus ten to ten with minus ten being absolute dictatorship and ten being perfect democracy. In the polity index more information is captured but the theoretical foundation becomes weaker (Bogaards, 2012). The next dimension in measuring is using either the level of democracy at one point in time or the country’s democratic history. From the following studies, it became clear that the strongest relations can be found using the democratic history opposed to the level at a single point in time. The reasoning here is that time is needed, when a country has switched regime type, for real change to materialize.
In this paper the focus is on income inequality and not on inequality of other kinds, such as gender, racial and religious inequality. In the next part of this paper the studies that presume a direct relation from democracy to inequality are under review. Government characteristics are a major determinant in a regime’s ability to increase or decrease inequality and are therefore the subject of the second section. Unambiguousness of causality will show to be problematic in finding conclusive evidence. Therefore the last two sections about general health and education relating to democracy provide an even broader context wherein inequality can be assessed. The literature review is followed by a conclusion about the research that has been done on democracy and inequality.
2
Democracy and Inequality
In this section, I lay out the studies and literature that have been conducted in the field of economics regarding democracy and inequality. I limited the time frame to the present and past century with the oldest paper I’m reviewing on the topic of inequality and democracy dating back to 1981. Most articles however are not more than 20 years old. Quite a lot has been said, proven and proven false in this period of time.
What are the effects of political democracy on economic growth and in-equality? Sirowy and Inkeles (1990) try to answer this question. Is democracy facilitating, hindering or is there no relation at all with development? From a theoretical point of view most economist agree, a facilitating effect of democ-racy on development makes the most sense. Sirowy and Inkeles instead find the level of political democracy in time not associated with lower levels of income inequality. Taking a new approach in the way results of past research can be interpreted, yields to the conclusion: democracy in time is not associated with lower levels of income inequality. In their meta-analysis of 12 studies, few robust relations were found.
In my opinion absolute robust results in economics can hardly ever be found due to the complexity of the world we live in. This doesn’t mean we should stop pointing out and try to fix econometric problems. (Muller 1988) lays out the problem of reverse causation from inequality to democracy. He shows that in cross national regressions of the Gini coefficient on democracy, econometric problems arise. Omitted variable bias is affecting both inequality and democ-racy. By using a larger dataset than most economists, Muller tries to increase the chance of solid conclusions. He confirms the strong inverse correlation be-tween income inequality and the political stability within a country. Democratic stability however seems independent of economic growth. Taking two different approaches in measuring democracy namely: level of democracy at a point in time or as democratic experience over time shows the correlation with income inequality is strongest when the ‘democracy as experience’ measure is used. Muller concludes democracy needs to be in place long enough for inequality to fall.
Despite Muller’s extended dataset, Weede (1989) still challenges the robust-ness of Muller (1988). An important critique is that replication with different data sets is not possible and adding a single case alters the outcomes significant. Also, the control variables used by Muller are claimed insignificant by Weede. He concludes: no clear judgment can be made on the relationship between democracy and income inequality. The critique from Weede seems valid but the question that remains unanswered is: when can we make a clear judgment?
The now following researchers all do make clear judgements about the sig-nificance of democracy on inequality but bring different views than Muller. Simpson (1990) on political rights and income inequality says the relation be-tween equality and national wealth is not linear but instead U shaped. Simpson remains vague about the idea behind this inverted polynomial U relationship. Democratic experience also does not have a direct effect on income inequality but may have an indirect effect. Still democratic rights and education are the best predictors and most important causes of income inequality. An important conclusion although the intuitions and causality remain vague. He does stress the strength and robustness of his findings.
Burkhart (1997) uses a simultaneous equation, where income distribution has an effect on democracy but democracy also influences income distribution. This makes sense because It could very well be that causality does not just go one way. Burkhart Believes in a downward spiral where lower democracy levels lead to income inequality, this income inequality then increases inequality between social classes which lowers democracy levels even further. But this works two ways. Low income inequality raises democratic levels and capitalism, while capitalism will increase capital inequality again. His theory on the topic comes down to: the spreading of political power in a democracy will spread economic power as well. Like with Muller (1988), there is no optimal stopping point for democracy, more is better.
Where Weede (1989) finds no evidence for Mullers claims, Huber et al. (2006) reproduced the findings of Muller (1989) and find results in favour of the relation between the democratic history and inequality. Their findings strongly support the alleged link between inequality and politics. Democracy matters for inequality in the long term because policies can be used to redistribute capital. Take in mind this is no guarantee but the probability of redistribution increases. Huber et al. (2006) use data from Latin America and the Caribbean, who as a whole have very high levels of inequality and varying regimes. The differences in regimes between countries in the Latin American region make this region useful for economic analysis that have a political element. Datasets on Latin America are thus often used in papers concerning my thesis subject.
Another wide used data series in the literature is the Polity measure for democracy. Rodrik (1999) finds a positive correlation between real wages in manufacturing and democracy using Polity III. He controls for many variables and finds robust evidence that the driving force behind the increase in wages are political competition and participation. The wage increase happens because democratic institutions shift the distribution of income from profits to wages. Institutions matter for the functioning of distribution. The most plausible hy-pothesis for explaining why institutions matter is that democracies allow for more efficient bargains than under authoritarian regimes who repress wages.
Rodrik found inequality decreases with democracy, Scheve and stasavage (2009) found inequality increases when democracy is lacking. Suffrage appeared negatively correlated with income accruing to the top 10-1 incomes. To come to this conclusion 13 OECD countries where analyzed from 1916 till 2000 and controlled for fixed effects. They used universal suffrage as a dummy. It has to be mentioned that no impact was found on the share of national income accruing to the top 1% wealthiest within the population. This is seems strange and the reason why no effect on the top 1% can be found remains unclear.
A whole different approach again comes from Li et al. (1998) who used an index of civil liberties and income inequality and found negative correlations. They give answer to the question how stable income inequality is within and between countries. Li et al. find relative stability of inequality over time instead of an inverted U shaped relation with income like Simpson (1990) earlier sug-gested. They find inequality changing slowly over time and therefore, the idea that political economy is a main force determining inequality seems plausible. That is given the believe political policy slowly changes overtime and not with humps and shocks. Nonetheless Li et al. do find democracy being not the most important determinant. According to their empirical analysis, imperfect capital markets have greater influence on inequality than regime type. Although
im-perfect capital markets are very interesting, the focus in the rest of this paper remains on regime types and their influence on inequality.
In order for a democracy to enable more equal income redistribution, the regime first has to collect tax. Cheibub (1998) researches taxation in cies and dictatorships. Cheibub wonders whether it is dictatorships or democra-cies that can collect more tax. To answer this question he looks at 108 countries for a period between 1970 and 1990. He finds that the differences between the level a country taxes are no result of regime type. His conclusion is not that regimes do not matter for economic policy but, that under some conditions they do and some conditions they don’t matter. The question that remains is: what are the conditions of influence making regime relevant?
Soifer (2013) is like Cheibub (1998) also focusing the importance of taxa-tion. In the article ‘state power and the economic origins of democracy’, Soifer stresses the importance of the ability for a government to create their own in-come in order to be able to redistribute. Suffrage expansion does not have to lead to increased redistribution if extractive capacity is lacking. Democracy and dictatorship are only part of the equation. When it comes to what facili-tates income redistribution, taxation is the most important measure regarding redistributive policy. If the government lacks the capacity to collect tax it does not matters whether the country is led by dictator or elected leader. Elites in power do not care for the type of regime in control if the state remains weak in terms of executing power. Many economist do stress the implications that come with the realisation democracy has many indirect effects on inequality. A good example is Lee (2005) in his article on income inequality, democracy and public sector size; Lee argues there are heterogeneous effects of democracy on inequal-ity. The varied effects are measured for interaction in a panel dataset consisting of 64 countries between 1970 and 1994. Significant interaction between size of government measured by tax revenue as percentage of GDP and democracy is observed. Public sector size and democracy affect income inequality. In non-democracies public sector expansion makes income equality worse. Democracy plays an important role in decreasing income inequality given that the public sector grows.
2.1
Democracy and Government
In the previous section the idea that democracy matters for inequality appeared multiple times. At the same time, the diverse nature of the implications a regime can have on the income distribution was largely ignored. In this part more emphasis is laid on the difference in government that comes with democracy and how this government then can address inequality. Policy is key.
Meltzer and Richard (1981) form the basic theoretic model for policy effects of democracy on a country. They conclude democracy increases tax revenue and income redistribution. The model is based on general equilibrium theory with rational utility maximizing individuals who make fully informed decisions about taxation and income distribution. In today’s postmodern world these kind of assumptions are made with more caution. Kuznets (1955) finds that economic growth increases incomes of skilled workers relative to unskilled workers. In line with Kuznets’s finding, Meltzer and Richard (1981) make the following prediction with their model. Economic growth that lead to rising inequality will increase votes for redistribution. This will then decrease the relative income gap
between skilled and unskilled workers.
Gil et al. (2004) puts these claims to the test in a comparison of 142 countries over the years 1960-1990. No correlation between tax revenue or government spending and democracy is found. No significant difference between democracies and non-democracies when it comes to economic and social policy appears. From this empiric study the role of democracy seems over empathised. Still there are many other studies that do find significant differences in policy between democracies and non-democracies. Gil et al. (2004) Is an outlier in finding no effects of democracy on tax revenue.
Where Meltzer and Richard (1981) reasoned starting from the neo-classic homo-economicus and built from there, Acemoglu and Robinson took a more considered approach regarding inequality, growth and development. Instead of reasoning from one totalitarian theory that tries to place every facet of the whole world in one big framework of economics, they search for a balance between empirics, theoretical reasoning and historical analysis.
Acemoglu and Robinson (2000) did an empiric study on the colonial origins of comparative advantages. They conclude differences in institutions over the years can explain three-quarters of income per capita differences across former colonies. The colonization strategy determent by the European settlers had a big influence on how the country later would develop. Colonies with better developed and advanced institutions would later reap the benefits in the form of stronger economic growth. Dictators fall and democracies rise followed by redistribution to the poor from the rich. Acemoglu and Robinson find rev-olutionary events a viable instrument for democracy. With panel data from western Europe dating 1820-1913 their conclusions appear robust. Democracy has a positive effect on government expenditure relative to GDP. Democrati-zation is argued to make credible commitment to redistribution and therefor takes away the threat of a revolution fuelled by inequality. This explains why moves towards democracy in response to social unrest are often followed by large increases in redistribution.
Another more historical study is done by Aidt and Jensen (2009b). For the period 1890-1938 a robust positive effect from increased suffrage to government expenditure and tax revenue is found in a sample of 10 western countries. Coun-terintuitive is, that suffrage expansion lowers direct tax and increases indirect taxation. The size of the government as a proportion of the economy in the developed world increased dramatically. While the introduction of income tax is linked to the extension of the franchise. The income taxation is then liked to demands for greater redistribution. Aidt and Jensen (2009a) find determi-nants of introduction of income tax having a non-linear relation with suffrage. First the likelihood of income tax is reduced but later when universal suffrage is implemented the likelihood of income tax increases.
Like Acemoglu and Robinson (2000), Huber and Stephens (2012) combine historical analysis, case knowledge and quantitative analysis in the form of panel data on Latin American countries from 1970-2007. Cumulative years since a country has been a democracy are regressed without fixed effects. Democracy appears significant positively correlated with education spending, healthcare spending, social security and welfare spending like we have seen many times before.
A complete different viewing angle comes from Scheve and Stasavage (2010, 2012). They again relativize the importance of democracy. Taking a long run
approach looking at OECD countries, Scheve and Stasavage find no correlation between democracy and tax progressivity or the rate of capital taxation. They find the rise of electoral democracy, being the driving force behind the politics of redistribution over empathised. Mass warfare (like WWI) plays a critical role in the development of redistributive public policies, they claim. Mass warfare increased demand to fairly distribute the burden of the war effort. In absence of war there is no force systematically associable with progressivity in tax. Their findings and emphasis on the importance of war align with Tilly (1985) and Besley and Persson (2011).
The warfare theory by Scheve and Stasavage (2010, 2012) finds some sup-port by Aidt and Jensen (2013). Growth in central government can be partly be explained with war finance hypothesis. Aidt and Jensen name wars as an important impulse to the public sector size growth. They move the literature forward by controlling for endogenous variables but more on that follows later. Someone that explicitly refutes war as leading force in redistribution is Lin-dert (1994). He uses panel data from the European Union, United States, Canada, Japan, Australia, Argentina, Brazil and Mexico from 1880-1930. And concludes the rise of social transfers through government was not led by post-war Germany. The rise in voting rights help explain income redistributions instead. He attributes the low level of social spending before the 20th century to the fact that the political voice was restricted. Lindert (1994) also marginalises the role of income growth. Income growth played less of a role in shaping the rise of social transfers than democracy did. His global view on welfare states and accent on importance of democracy inspire new ideas in the field.
A more nuanced conclusion, comes from Brown and Hunter (1999) asking: what the effect of regime type on public expenditures is for social programs. Brown and Hunter too look at Latin American but in a slightly different time period namely 1980 and 1992. Some evidence is found to indicate democracies spend more on social welfare than autocracies when controlling for GDP and other macro-economic figures. They remain very inconclusive, something that makes reading their piece unfulfilling in my opinion. On the other hand, the complexity of the subject is recognised and the temptation to jump to conclu-sions is repelled. Governments can increase or decrease social spending and there is no one country characteristic that determines social spending under all circumstances.
Another study concerning the effects of democracy in Latin America comes from Kaufman and Segura-Ubiergo (2001). Using IMF data of fourteen Latin American countries from 1973 to 1997. They find a positive correlation between democracy and government, health and education spending. The effects of democracy on health and education are addressed later in this paper. They conclude Latin America’s segmented labour markets strengthens inequality and this is named as a reason why democracy works systematic to hold down social security spending. Nonetheless, in general Latin American countries support progressive forms of social spending. Questions about causal mechanisms remain unanswered once again.
An explanation why progressive social spending increases with democracy is by some attributed to the expansion of the female franchise. The first example is Aidt and Dallal (2008). They find that from 1869-1960 the major driving force in the increase of progressive social spending comes from the female vote. Like Lindert (1994) argued, they confirm female enfranchisements independent
effect on social spending. With a short-run effect that is significant and a long long-run effect even more convincing. The fiscal and economic consequences of universal suffrage generally led to considerable increases in social welfare, educational, and housing spending.
The second example comes from Lott and Kenny (1999) for the period 1870-1940. Also studying the impact of female votes on redistributive policy. The paper begins with the statement: women vote differently than men. In the paper woman are seen as more risk averse, and this is the reason for the increase in so-cial security spending after woman enfranchisement. Woman suffrage increased per capita state revenue and expenditure. They reason, women vote to secure themselves of income in case they would lose their man or remain single. This incredible patriarchal reasoning and world view is supported with a correlation but it proves that anything goes when it comes to explaining economics based on regression. There is an issue with causality: there could be something else that increases both woman suffrage and larger government at the same time.
The third paper that links female franchise to progressive social spending in again another way is Miller (2008). He argues women’s votes helped child welfare improve more than those of men. Infant mortality fell with the introduction of voting rights for women. Public health spending increased and child mortality declined by around 10%. Children were able to benefit more fully from scientific breakthroughs through increased women voting rights. Allowing women to vote has had a positive effect on government policy regarding healthcare and social spending according to the last three papers.
For me the most convincing evidence that democratization increases the size of government comes from Aidt and Jensen (2013). Finding central government has expanded from less than 5% of GDP to about 20% of GDP for most western countries in the 19th century. They use an identification strategy to tackle the factor democracy as endogenous. They are one of the view that address econometric problems concerning democracy. In general the literature does not recognize or tackle difficulties with econometric problems like correlated effects, variable bias and selection bias.
2.2
Democracy and Health outcomes
At the end of the previous section the growth of government as percentage of GDP as a result of democratization and its effect on inequality became clear. Miller (2008) then found a robust link between the increase of the franchise and child mortality decline. In this part the emphasis is completely on the effect of democratization and health inequality.
Strong conclusions on this topic come from Mcguire (2010). In his book Wealth, Health and Democracy he argues that in East Asia and Latin America: healthcare and other social services that reduced mortality have been achieved through political democracy. Based on case studies of Latin America and East Asia. By now we have seen that there are incredible differences in inequality be-tween political regimes. The idea ‘wealthier is healthier’ makes sense within this frame work. There should be an extend of health differences between regimes if inequality differs. Mcguire (2010) indeed finds democracy positively correlated with life expectancy.
Besley and Kudamatsu (2006) do not believe in evidence for democracy pro-moting prosperity or wealth. These correlation can’t be strong nor robust they
claim. In a cross country panel dataset from 1960-2000, Besley and Kudamatsu do find a robust correlation between democratic institutions and health. With again greater life expectancy and lower infant mortality in democracies despite the lacking wealth effect. Political institutions seem to matter for health in-equality in a direct way and not via more equal wealth distribution.
More evidence that democratic societies healthier than autocracies comes from Wigley and Akkoyunlu-Wigley (2011). Using a panel of 153 countries for the period 1972-2000, they test if democratic governance alone has a positive influence on health. Where Health is measured as life expectancy. The theory is that: in democratic countries elected representatives can be held accountable by voters if policies fail to treat ill health. The democratic experience of a country does have a positive effect on life expectancy. The effect remains when controlled for the distribution of health promoting resources. So political institutions thus provide only a partial explanation on the impact of regime on health.
A new approach by Blayeds and Kayser (2011) links calorie consumption instead of life expectancy to well-being and equality in democracy. Their study is all based on the Polity IV database and calorie data from the Food and Agriculture Organization of the United Nations. Calories consumed decline rapidly with increase of income. Thus, calorie consumption is taken as a proxy for poverty. They show, given similar rates of economic growth, democracies make more food available measured in total calories. At the same time appears democracy best for the poor and pro-poor policies. One of their explanations is that democracy is good for manufacturing wages and investments in human capital. This kind of creative research design, although rare in this field, is what truly brings a valuable addition to the existing literature.
As from other studies reviewed appeared, also Gerring et al. (2012) shows history of democracy yields stronger correlations than democracy at a point in time. Using data from 1960-2000 and also a polity score, no robust or strong correlation between regime type in a country and infant mortality is observed. However, there is a robust causal relationship if looked at history of democracy. This makes sense because there are strong theoretical grounds for expecting a causal relationship to exist between human development and the history of democracy. In longer existent democracies there is time for political competition to develop, also leading to greater accountability for the executing power.
The whole idea that democracy provides better health equality is doubted by Ross (2006). In the past decades general health has increased and in many countries prevalence of democracy increased. It could be that there is a direct effect from democracy to health equality. It could be that democracy increases wealth and then increases health. Ross believes democracy unquestionably pro-duces non-economic benefits for people in poverty, but few if any improvements in their material well-being. His explanation for finding benefits of democracy in other studies could be due to selection bias that led to overestimating bene-fits of democracy. Like we have seen before regarding effects of democracy on inequality: the truth remains unclear.
2.3
Democracy and Education
In this final section literature investigating education and democracy is under review. Economists describe personal growth and education often in terms of human capital. Human capital theory is widely used to explain wage differences
so it makes sense that education could affect income inequality. In simplified terms one could say: more education leads to a higher wage. So if in democracy access to good education is easier and more widely available, income inequality is expected to be lower.
Exactly this intuition is confirmed by Lindert (2004) in his book Grow-ing Public. The emergence of democracy increases education spendGrow-ing. UsGrow-ing historical insights, political analysis and economic assessment he comes to the conclusion that the spread of democracy played a major role in the rise of so-cial expenditures. Another important insight from his book is the increase in social expenditure not decreasing incentives to self-provide as an individual. In general is government growth positive for society says Lindert and education spending plays a big role reducing inequality.
A more indirect link to inequality reduction via education from Engerman and Sokoloff (2005) is based on a historic approach looking into institutions. Europeans established colonies over the world with greater equality among the population tended to extend the franchise earlier and more broadly. This char-acterization is measured between countries and within the United States. Soci-eties with greater equality or homogeneity tended to adopt suffrage institutions that again provided broader suffrage. How suffrage evolved appeared signifi-cant for long run patterns of economic development. In the USA they find a close connection between education and the extend of democracy. This study has great similarities to Acemoglu and Robinson (2000) in the way it explains observations with the history of European colonies.
Finding systematic effects of democracy on domestic policy has been hard using a historic approach. Baum and Lake (2001) managed to develop a theory of the state similar to standard economic theory’s, suggesting similarities be-tween states, firms and monopolies. Baum and Lake conclude that democracy has real effects. Their data finds public services increase under democracy. For instance secondary school enrolment increases across the developed world with increasing democracy. They conclude democracy works.
With a less solid theoretical foundation but thorough empirical research, Brown and Hunter (2004) find a positive correlation between total education spending per capita and the polity index. The dataset consists of 17 Latin American countries dating from 1980-1997. Democracies spend more on edu-cation and therefore provide better prospects in human capital formation espe-cially due to better primary schooling. They remark however that the greater spending does not per definition yields greater learning. Something that has to be investigated more carefully.
In an earlier study by Brown (1999) he also examines the relationship be-tween regime type and human capital with a slightly different approach. He mirrors the dichotomous measure created from the Polity dataset. In this study Brown recognizes that finding causal links between democracy and inequality or education is hard because the wide array of complex institutions. It’s not hard to point out potential links, but it is hard to point out the ones really of rel-evance. Despite this precaution Brown again stresses the relationship between democracy and accumulation of human capital. Given educations importance for economic growth, democracy appears important to.
Stasavage (2005a) uses a similar research design compared to Brown and Hunter (2004). The main difference is the geographic area. Stasavage (2005a) looks at sub Saharan Africa. Stasavage also uses the impact of democracy
on education now measured by Przeworski’s dichotomous measure. Relatively little research focuses on Africa, but similar to studies on other continents he finds democracy increasing total education spending as percentage of GDP. He stresses the importance of electoral competition as a force in influencing providence of basic services. Even though institutions in African countries are considered to be weak, the conclusion is change in economic policy is the result of democracy.
The same year Stasavage (2005b) publishes a more unique study on Uganda and its decision to move to universal primary education. He shows why in Uganda primary school expenditure grew but also why democratic transitions in other African countries might not have the same effect. The successfulness of improving African governments conditioned on two factors. The first is broad access to information about government policy. The second factor is the candi-dates who must choose to make public services part of their campaign strategy. If one of the two factors is not fulfilled a democratic transition won’t reduce inequality predicts Stasavage.
Later Harding and Stasavage (2013) reconfirm the impact of democracy on primary education in sub Saharan Africa. Now looking at primary school en-rollment on a national and individual level. Again Stasavage and Harding find democratic governments more likely to abolish primary school fees. If a policy measure is easily observed, the likelihood of the measure being implemented increases. Promises for hiring more teachers simply is less easy to verify then abolishing school fees for instance. This evidence for the importance of infor-mation builds on the initial theory by Stasavage (2005b).
The strong and extensive evidence by Stasavage can be put into context by Gallego (2010). He is one of the few who attempt to develop an identification strategy for determining the impact of democracy on education. He tackles the issue of reverse causation. Using the same data as Acemoglu (2001) on Settler mortality from 1500-1900. Polity scores appear to do have a significant causal effect on primary school enrolment. Differences in schooling attainment are related to political institutions which are affected by colonial factors. But he argues educational institutions are endogenous and therefore costly to change. Gallego concludes the effect of enfranchisement on education might be overrated. The concern for reverse causation relevant for almost all studies in this pa-per that was brought up by Muller (1988), is something Gallego claims to solve. Acemoglu et al. (2005) also reduces the concern of reverse causation for edu-cation and democracy. He states that claims of high levels of schooling being both necessary for, and a result of democracy, not robust. The cross-sectional reverse relation between education and democracy is mostly driven by omitted variables.
In the book From the Ballot to the Blackboard by Ansell (2010) it becomes clear that the positive effect of public education spending can be offset by a reduction in private education spending. Ansell looks at a broad sample of 100 countries from 1960 till 2000. Ansell instruments for democracy and finds democracy having a positive significant effect on educational spending as per-centage of GDP. But there is also a negative correlation between democracy and private education spending as percentage of GDP. Still the on average total education spending increases but the positive impulse to education is not one dimensional.
and education. Aghion et al. (2012) surprisingly find a negative correlation. Based on data from the last 150 years, they argue that primary education was influenced most by military rivalry. Democratic transitions are negatively as-sociated with education investments. This goes against most other research and the majority of their conclusions. Aghion et al. give an explanation how this conclusion can coexist with other research. The presence of political in-stitutions does positively affect military rivalries. This study again proves that correlations do not tell us much.
3
Conclusion
My aim in this paper is to provide a clear overview of different research perspec-tives in the field of economics concerning the relation of democracies to income inequality throughout the years. Most studies do find the level of democracy in time associated with lower levels of income inequality. Whether this relation is strictly positive, linear or inverted U shaped is discussed however most re-searchers do agree the relation is positive. A big and returning issue in finding a relation between democracy and income inequality is robustness. By expanding the scope of research to indirect effects more data becomes available to tackle this problem.
In the following section on government size, the conditions necessary for a democracy to succeed in bringing down income inequality become clear. The conclusion is that it does not matter whether a country is a democracy or not, change can only be realized if the government has the power to install the necessary political institutions. When regime power is strong enough, many other country specific characteristics remain to determine if social spending rises or falls. Government growth does often results in higher spending in healthcare and education. Different theories on how healthcare and education spending are closely related to income inequality pass by. In the case of health outcomes strong conclusions fade because of the many possible explanations for improved health in the population besides regime type (Ross, 2006).
In case of education spending the theoretical foundation for believing more education spending would lead to greater income equality is strong. With the assistance of human capital theory, clear conclusions are made about the role of education for income. The successfulness of education, with success in terms of knowledge, is hard to define and this problem remains largely undiscussed in the literature. The same problem concerning robustness that was found in the first section of this paper returns. Even the repeatedly found positive correlation between education and democracy is arguably negative by Aghion (2012).
The main conclusion from all reviewed journal articles in aggregate is: there appears to be a positive correlation between democracy and inequality but this relation is precarious. Causation however is not convincing at all and finding a causal relation from democracy to lower inequality is time after time harder than expected. Econometric techniques can only do so much in laying bare true economic relations. The complexity in the different factors that play a role in determining inequality make it hard to point out if correlations are meaningful or not.
Most studies on democracy and inequality have employed similar research designs. The research methods used in this field are also surprisingly similar, with some exceptions. Almost all articles follow the structure: introduction, literature, data, results, conclusions. More than once in going over the eighty papers initially selected, I felt like reading more or less the same article. Re-freshing papers with new perspectives and theoretical insight were sometimes hard to find but in the end a broad scala of perspectives on the problem of democracy and inequality passed by. Many claim their own results to be more robust than others but this of course can not be true for all studies. I found that correlation based on simple regression for a period of thirty years almost never gives valuable insight. Lacking causal inferences make it impossible to demonstrate the true relevance of democracy on inequality.
The world is more complex than often recognized. It is very hard to find conclusive evidence. But creativity is key in tackling this problem. Robustness is an inherently vague principle in the context of democracy and therefore no consensus will be reached on when it is achieved or when it is not. Many factors determining income distributions are at play and these are also differing between countries. Churchill said democracy is terrible but still the best option that we have tried. From an income inequality perspective I would say: democracy seems to be doing okay for a lot of countries, however it certainly does not have to work for the entire world.
References
Acemoglu, D., Johnson, S., Robinson, J. A., & Yared, P. (2005). From educa-tion to democracy? (Tech. Rep.). Naeduca-tional Bureau of Economic Research.
Acemoglu, D., & Robinson, J. A. (2000). Why did the west extend the fran-chise? democracy, inequality, and growth in historical perspective. The Quar-terly Journal of Economics, 115 (4), 1167–1199.
Aghion, P., Persson, T., & Rouzet, D. (2012). Education and military rivalry (Tech. Rep.). National Bureau of Economic Research.
Aidt, T. S., & Dallal, B. (2008). Female voting power: the contribution of women’s suffrage to the growth of social spending in western europe (1869– 1960). Public Choice, 134 (3-4), 391–417.
Aidt, T. S., & Jensen, P. S. (2009a). The taxman tools up: An event his-tory study of the introduction of the personal income tax. Journal of Public Economics, 93 (1), 160–175.
Aidt, T. S., & Jensen, P. S. (2009b). Tax structure, size of government, and the extension of the voting franchise in western europe, 1860–1938. International Tax and Public Finance, 16 (3), 362–394.
Aidt, T. S., & Jensen, P. S. (2013). Democratization and the size of govern-ment: evidence from the long 19th century. Public Choice, 157 (3-4), 511–542.
Ansell, B. W. (2010). From the ballot to the blackboard: The redistributive political economy of education. Cambridge University Press.
Besley, T., & Kudamatsu, M. (2006). Health and democracy. The American economic review , 96 (2), 313–318.
Besley, T., & Persson, T. (2011). Pillars of prosperity: The political economics of development clusters. Princeton University Press.
Blaydes, L., & Kayser, M. A. (2011). Counting calories: Democracy and distribution in the developing world. International Studies Quarterly , 55 (4), 887–908.
Brown, D. S. (1999). Reading, writing, and regime type: Democracy’s impact on primary school enrollment. Political Research Quarterly, 52 (4), 681–707.
Brown, D. S., & Hunter, W. (1999). Democracy and social spending in latin america, 1980–92. American Political Science Review , 93 (04), 779–790.
Brown, D. S., & Hunter, W. (2004). Democracy and human capital forma-tion: education spending in latin america, 1980 to 1997. Comparative Political Studies, 37 (7), 842–864.
Burkhart, R. E. (1997). Comparative democracy and income distribution: Shape and direction of the causal arrow. The Journal of Politics, 59 (1), 148– 164.
Cheibub, J. A. (1998). Political regimes and the extractive capacity of gov-ernments: Taxation in democracies and dictatorships. World Politics, 50 (03), 349–376.
Engerman, S. L., & Sokoloff, K. L. (2005). The evolution of suffrage institutions in the new world. The Journal of Economic History, 65 (04), 891–921.
Gallego, F. A. (2010). Historical origins of schooling: The role of democracy and political decentralization. The Review of Economics and Statistics, 92 (2), 228–243.
Gerring, J., Thacker, S. C., & Alfaro, R. (2012). Democracy and human development. The Journal of Politics, 74 (1), 1–17.
Harding, R., & Stasavage, D. (2013). What democracy does (and doesn’t do) for basic services: School fees, school inputs, and african elections. The Journal of Politics, 76 (1), 229–245.
Huber, E., Nielsen, F., Pribble, J., & Stephens, J. D. (2006). Politics and inequality in latin america and the caribbean. American Sociological Review , 71 (6), 943–963.
Huber, E., & Stephens, J. D. (2012). Democracy and the left: social policy and inequality in latin america. University of Chicago Press.
Kaufman, R. R., & Segura-Ubiergo, A. (2001). Globalization, domestic politics, and social spending in latin america: a time-series cross-section analysis, 1973– 97. World politics, 53 (04), 553–587.
Lake, D. A., & Baum, M. A. (2001). The invisible hand of democracy: political control and the provision of public services. Comparative political studies, 34 (6), 587–621.
Lee, C.-S. (2005). Income inequality, democracy, and public sector size. Amer-ican Sociological Review , 70 (1), 158–181.
Li, H., Squire, L., & Zou, H.-f. (1998). Explaining international and intertem-poral variations in income inequality. The Economic Journal , 108 (446), 26–43.
Lindert, P. H. (1994). The rise of social spending, 1880-1930. Explorations in Economic History, 31 (1), 1–37.
Lindert, P. H. (2004). Growing public: Volume 1, the story: Social spend-ing and economic growth since the eighteenth century (Vol. 1). Cambridge University Press.
Lott, J. R., Jr, & Kenny, L. W. (1999). Did women’s suffrage change the size and scope of government? Journal of political Economy, 107 (6), 1163–1198.
McGuire, J. W. (2010). Wealth, health, and democracy in east asia and latin america. Cambridge University Press.
Meltzer, A. H., & Richard, S. F. (1981). A rational theory of the size of government. Journal of political Economy, 89 (5), 914–927.
Miller, G. (2008). Women’s suffrage, political responsiveness, and child survival in american history. The Quarterly Journal of Economics, 123 (3), 1287–1327.
Muller, E. N. (1988). Democracy, economic development, and income inequal-ity. American Sociological Review , 50–68.
Mulligan, C. B., Gil, R., & Sala-i Martin, X. (2004). Do democracies have different public policies than nondemocracies? The Journal of Economic Per-spectives, 18 (1), 51–74.
Rodrik, D. (1999). Democracies pay higher wages. The Quarterly Journal of Economics, 114 (3), 707–738.
Ross, M. (2006). Is democracy good for the poor? American Journal of Political Science, 50 (4), 860–874.
Scheve, K., & Stasavage, D. (2010). The conscription of wealth: mass warfare and the demand for progressive taxation. International Organization, 64 (04), 529–561.
Scheve, K., & Stasavage, D. (2012). Democracy, war, and wealth: lessons from two centuries of inheritance taxation. American Political Science Review , 106 (01), 81–102.
Simpson, M. (1990). Political rights and income inequality: A cross-national test. American Sociological Review , 682–693.
Sirowy, L., & Inkeles, A. (1990). The effects of democracy on economic growth and inequality: A review. Studies in Comparative International Development , 25 (1), 126–157.
Soifer, H. D. (2013). State power and the economic origins of democracy. Studies in Comparative International Development , 48 (1), 1–22.
Stasavage, D. (2005a). Democracy and education spending in africa. American journal of political science, 49 (2), 343–358.
Stasavage, D. (2005b). The role of democracy in uganda’s move to universal primary education. The Journal of Modern African Studies, 43 (01), 53–73.
Tilly, C. (1985). War making and state making as organized crime. Violence: A reader , 35–60.
Weede, E. (1989). Democracy and income inequality reconsidered. American Sociological Review , 54 (5), 865–868.
Wigley, S., & Akkoyunlu-Wigley, A. (2011). The impact of regime type on health: does redistribution explain everything? World Politics, 63 (04), 647– 677.
