VoxEU Column Development Institutions and economics Politics and economics

The value of democracy in the world’s poorest region: Evidence from Kenya’s road building

Ethnic favouritism is a longstanding problem in Africa. This column presents new evidence of this phenomenon and how democracy affects it. Data on road building in Kenya confirms strong ethnic favouritism that disappears during periods of democracy.

An enormous literature points to a diverse set of factors behind Africa’s growth tragedy, ranging from bad policies, poor education, and poor infrastructure, to aging leaders, the historic slave trade, and political instability. Historians, political scientists, and economists have all argued that ethnic favouritism – a situation where coethnics benefit from patronage and public policy decisions – has hampered the economic performance of many African countries. In particular, scholarly work by Easterly and Levine (1997) attributes much of the poor economic performance seen in Africa up to the end of the 1980s to political frictions and economic mismanagement associated with ethnic fractionalisation. According to several accounts, ethnic favouritism ultimately emerges because weak political institutions are unable to constrain governments from discriminating among citizens (e.g. Bates 1983, Horowitz 1985, Esman 1994, Mamdani 1996). To understand the recent political and economic performance of many African countries – where the growth rate per capita has been positive since the 1990s after nearly three decades of negative growth (see Figure 1) – this provides an opportunity to determine to whether ethnic favouritism is still prevalent, and whether the emergence (or in many cases, re-emergence) of democracy has helped mitigate it.

Despite popular sentiments Africa is indeed becoming more democratic (see Figure 1).1 Between the early 1960s (when many African countries gained independence from colonial rulers) and the end of the Cold War in 1991, not a single African ruler was peacefully succeeded at the ballot box, except in Mauritius. But since Benin’s Mathieu Kérékou and Zambia’s Kenneth Kaunda bequeathed power graciously in 1991, more than 30 African leaders or ruling parties have let their countries’ voters decide. Multiparty systems in Africa now far outnumber single-party ones. This contrasts strikingly with the Arab world, where so far almost no incumbent-ejecting elections have taken place anywhere.

Figure 1.

In our recent paper Burgess, Jedwab, Miguel, Morjaria and Padro i Miquel (2013) we revisit history and try to credibly quantify the extent of ethnic favouritism in public expenditure in Kenya for the post-independence period. We examine whether the transition into and out of a multiparty democracy affects the extent of ethnic favouritism.

Empirical challenges
  • Obtaining data on public expenditure in Africa is challenging due to underfunded statistical agencies; when data does exists, it is patchy, of poor quality, and rarely disaggregated, therefore making it impossible to uncover which ethnic group is getting what share of public expenditure.
  • Estimation of ethnic bias requires observing what happens with public expenditure when there are switches of the ethnic group in power. In many African countries this is difficult given the long tenures of post-independence leaders and the fact that particular ethnic groups have tended to be dominant for extended periods.
  • To estimate the impact of institutional changes such as democratisation on ethnic favouritism, one would ideally observe switches between democracy and autocracy under the same leader, which is far from common.
Our approach

We overcome several of the challenges:

  • Post-independence Kenyan districts are dominated by a particular ethnic group.2 This geographic pattern of ethnic demographics has remained stable over time. This allows us to directly assess, using road outcomes by district, whether or not ethnic groups that shared the ethnicity of the president disproportionately benefited from roads.
  • Road expenditure can be directly measured. We use historical project level data to recover road expenditure data. This has enabled us to construct district-level panel data on road expenditure and actual road construction for all 41 Kenyan districts across the entire post-independence period of 1963-2011.
  • In the study period, 1963-2011, roads are the largest single element of public expenditure in Kenya (at 15%). This is three times the government expenditure on health, education or water. Road expenditure is centrally allocated and is a highly visible form of public investment and thus a prime target for ethnic favouritism.

The post-independence history of Kenya provides us both with switches in the ethnicity of the president, and switches into and out of multiparty democracy under the same president as illustrated in Figure 2.

Figure 2.

Our findings

We first illustrate our results using maps of the road network across key years in Kenya’s political history (see Figure 3). The pink area represents the six Kalenjin dominant districts (11% of national population), while the green area represents seven Kikuyu dominant districts (20% of national population). Comparison of the paved road maps for 1979 and 1992 gives a first visual indication of ethnic favouritism. The paved road network in and around the green Kikuyu districts appears largely frozen between these two years. In contrast, the paved road network in and around the pink Kalenjin areas expands dramatically. Conversely, after 1992 the expansion in and around the green and pink areas is much more equal.

Figure 3.

We next construct a summary statistic to detect favouritism: the share of road expenditure received by a district out of the total national road development budget that year, divided by the national population share of the district. A value of one indicates a district received road spending that is exactly proportional to its population, while values greater than one indicate spending that is above the national average.

We splice districts in two ways. First by whether or not, in a given year, the majority ethnic group in a district (50% and above) is the same ethnicity as the president. This allows us to examine whether or not districts that are coethnic with the president received a higher share of national spending on roads relative to their share in the national population.

Figure 4a.

Figure 4a plots the evolution of the summary statistic described above.3 Two interesting patterns emerge:

  • During periods of autocracy (the 1970s and 1980s) the ratio of district share of road expenditure to district share of population is always above one for coethnics and below one for non-coethnic districts, which is a strong indication of ethnic favouritism; and
  • During multiparty democracy (1960s, 1990s and 2000s) the ratio is consistently lower and tends to be near one on average for both types of districts, implying little favouritism. Note that this pattern is within leadership.

Next we categorise districts into majority Kikuyu, Kalenjin (the two ethnic groups that produced presidents) or ‘other’ district category. With these three types of districts we repeat the figure above to ensure that the pattern observed in Figure 4a is not driven by a particular ethnic group (see Figure 4b). The results are telling – Kikuyu districts receive road spending in line with their population share during the democratic years. Following the banning of opposition political parties in 1969, road spending concentrates in these districts and rises to twice their population share until the death of Kenyatta (1978). After which we see a meteoric rise in Kalenjin districts allocation, and at the same time a decline in Kikuyu districts’ allocation. Kalenjin districts continue to obtain more than their population share until the re-introduction of multiparty democracy (1992). After 1992 we see a diminished effect on coethnic districts but a revival of the other districts. For the other districts – those that never had a coethnic president – we observe a U-shaped pattern in their allocation. Democracy seems to have a levelling influence in ensuring that Kenyan districts receive roads resources roughly in line with their population share irrespective of their ethnic alignment with the president. The stark pattern in the road investments we observe in presidential coethnic districts as Kenya enters and exists autocracy is extremely difficult to reconcile with spending being driven by economic efficiency.

Figure 4b.

Across 1963-2011, Kenyan districts that share the ethnicity of the president receive twice as much expenditure on roads.4 This is unequivocal evidence of an extreme degree of ethnic favouritism. However, these biases are not constant. While in periods of autocracy, coethnic districts receive three times the average expenditure in roads and five times the length of paved roads, both these biases disappear entirely during periods of democracy. Thus, the political regime is an important determinant of ethnic favouritism.


Closer examination of recent Kenyan history sheds light on how the re-emergence of multiparty democracy in the 1990s changed the nature of constraints on Kenyan leaders and altered the allocation of public resources. Multiparty democracy heralded an increase in mass political participation and lessened constraints on popular expression. There was a reduction in press censorship, and an explosion of private print and electronic media. These changes led to far greater scrutiny of the actions of executive authorities starting in the 1990s, which helps explain why ethnic favouritism dramatically reduced during periods of multiparty democracy.

Despite slow progress Africa is becoming more democratic. Twenty years ago, only a few African governments held proper elections. Now, almost all do. It is up to Africans to demand representative governments; a challenge, given that nearly half the continent lives on $1.25 per day.


Easterly and Levine (1997), “Africa’s Growth Tragedy: Policies and Ethnic Divisions”, The Quarterly Journal of Economics, 112 (4) 1203-1250.

Burgess, Jedwab, Miguel, Morjaria and Padro I Miquel (2013), The Value of Democracy: Evidence from Road Building in Kenya, Manuscript, LSE.

Bates, R H (1983), “Modernization, Ethnic Competition, and he Rationality of Politics in Contemporary Africa”, in D Rothchild and V A Olunsorola (eds.) State versus Ethnic Claims: African Policy Dilemmas, Boulder, CO: Westview Press. 

Horowitz, D L (1985), Ethnic Groups in Conflict, Berkley: University of California Press.

Esman, M. (1994) Ethnic Politics, Cornell University Press.

Mamdani, M (1996), Citizen and Subject: Contemporary Africa and Legacy of Late Colonialism, Princeton University Press.

1 Democracy is measured by the Polity IV measure compiled by political scientist at University of Maryland. Polity IV defines three regime categories: autocracies (-10 to -6), anocracies (-5 to +5) and democracies (+6 to +10). See http://www.systemicpeace.org/polity/polity4.htm

2 A result of colonial era border delineation.

3 The solid lines 1969 and 1992 indicate the transition away from multiparty democracy and back to multiparty party democracy. The broken lines in 1979 and 2002 capture presidential transitions.

4 We repeat the exercise using the length of paved roads and find that four times the length of paved roads are built relative to what would be predicted by their population share. Results are presented in a regression framework and are robust to different normalisations and measures of coethnicity; see Burgess, Jedwab, Miguel, Morjaria and Padro i Miquel (2013) for details.

1,050 Reads