Financial development and the collapse of clientelism in Ghana

In this research we develop a new hypothesis about the relationship between clientelism and financial development, and we will test it using data from Ghana. In large parts of the developing world clientelism operates between political parties and firms. Firms bankroll increasingly expensive election campaigns, and in return political parties offer and promise clientelistic favors (such as government contracts, exclusive privileges, monopolies etc.). Such a clientelistic equilibrium involves firms sorting into supporting different parties, perhaps based on ethnicity. Supporting a party with campaign finance is a gamble, if your party wins you do well, if they don’t you lose out. Such a situation emerged in Ghana after the democratization of the country in the 1990s. In this research we will examine how the extent of financial, particularly banking sector, development influences the preferences of businesspeople for clientelism. The hypothesis is that financial development creates a common interest between businesspeople which previously did not exist. This happens because if there is a change of power at an election and the winning political party reneges on contracts given by the old government, the negative impact on the financial sector threatens to bring down the entire private sector, including the “winners” at the election. Thus we hypothesize that deepening financial intermediation should reduce firms demand for clientelism. Ghana is an ideal place to test this hypothesis because since 1990 it has seen both a transition to democracy with rotations in power and a very large expansion of the financial and banking sectors. We will collect survey data from 200 firms on support for and donations to political parties, the extent to which the firms receive government contracts and their attitudes towards clientelism. The main explanatory variables will be the extent to which a firm is connected to the financial system, extent of deposits (which a firm may lose if the banks go under) and loans and other financial assets. We will also collect a lot of other data as covariates, such as economic sector, capital stock etc. The hypothesis can be looked at in cross-section: the more exposed a firm is to the financial sector, the less it engages/likes/approves of clientelism. We also hope to examine the within variation by asking firms retrospective questions so that we can try to examine whether as a firm becomes more exposed to the financial sector, it engages/likes less clientelism.