A couple of months ago, my friend and colleague Martin Rode (University of Navarra) and I put up a new version of our database. The data are an update and expansion of the much-used DD dataset, compiled by Cheibub, Gandhi and Vreeland; the data are available here. It, for example, includes three different indicators of democracy – all based on a highly minimalist conception of democracy – additional details of countries’ political institutions, and information of 542 coup attempts. The data are available on a yearly basis for 208 countries between 1950 and 2022.
One of several questions that one can use the database to answer is if the political institutions implemented by colonial powers continue to shape modern political institutions after the countries became independent. Martin and I decided to have a closer look at that exact question, which resulted in a paper that I presented at the 2019 Australasian Public Choice Conference in Brisbane, and also presented at an online seminar for Economic Research South Africa; the entire seminar can be viewed here. The paper was published this summer in the Journal of Institutional Economics.
In the paper “Late colonial antecedents of modern democracy”, we use one of the features unique to our database that Cheibub et al. did not offer: We have information on the structure of the political institutions of about 90 countries that were colonies in 1950. These institutions varied substantially: While Djibouti had a Representative Council in 1950, consisting of 10 nominated members, 10 elected from the first (White) college, and 10 from second (African) college, Rhodesia had a fully elected Legislative Assembly of 30 elected members. In the French colony of Djibouti, the representative institutions were biased directly by design while those of Rhodesia did not formally descriminate against Africans. However, Rhodesian democracy was unfair as a literacy restriction on voting effectively excluded about 85 percent of Black Rhodesians.
Our results can be subsumed in the simple table below, where we plot how many countries were non-democratic / democratic in the years immediately before independence against how many are now democratic or not. We find that if the colonial authorities implemented actual democracy prior to independence – and kept their fingers off the elections – it is highly likely that the country is a democracy today. Of the 39 countries in our sample that were democratic before independence, 30 of them are democratic today. Only 7 are democratic today without having had democratic political institutions during the colonial period.
|Colonial autocracy||Colonial democracy|
|Modern autocracy||29 (18)||9 (6)|
|Modern democracy||7 (6)||30 (27)|
In recent years, there has been a vigorous debate about the effects of colonialism and whether colonialism was all bad or good. Our findings indicate that in some cases, colonialism left a lasting, positive legacy in the shape of a viable democracy. One of the interesting details that are obvious in our database is that the British and Dutch colonial authorities were far more likely to implement actual democracy than the French, Portuguese or Belgian. In particular, we find that although the French introduced la Loi Cadre in 1956 that required the introduction of democratic institutions in all colonies, they rigged or otherwise managed the elections in almost all colonies; Dahomey (modern-day Benin) appears to have been the only exception.
That said, it didn’t always work out well for the British: Democracy proved short-lived in Sudan and the budding democracy in Rhodesia – which would have become a full democracy in due time because the vast majority of Black children went to school in the 1950s – was dismantled by Ian Smith’s appartheid government. The association is not perfect, but colonial democracy has turned out to be surprisingly long-lasting