In many African countries mobile phones outnumber adults, with the near universal coverage allowing direct communication with entire populations. A new study in the Proceedings of the National Academy of Sciences shows mobile technology, when combined with information on corruption and budget mismanagement, presents an opportunity to improve democratic outcomes.
Citizens should be able to vote out corrupt politicians; yet in many developing democracies this is not easy. Voters often lack sufficient information and capacity to punish corrupt politicians due to state-control of the media, low civic education, untrustworthy institutions, and uncompetitive elections.
Uganda, a landlocked country in East Africa, suffers from many of these challenges. Ugandans’ perceptions of corruption in their own country are amongst the highest in the world, while many of the former British colony’s democratic institutions are fragile.
The country’s reputation took a knock during the controversial 2016 election, where incumbent president Yoweri Museveni extended his 30-year rule after his opponent was arrested during the campaign.
These factors, alongside Uganda’s enthusiastic embrace of mobile phone technology, made the country an ideal case for Dr Ryan Jablonski of the Department of Government, and co-authors Dr Mark Buntaine, Dr Paula Pickering and Dr Daniel Nielson, to test whether technological interventions can improve democratic outcomes.
Dr Jablonski said that many Ugandan voters have a high level of mistrust of their political system. He noted, “Many voters are suspicious of information, which may be influenced by the state to conceal corrupt activities by public officials. This means they often don’t feel able to hold politicians properly accountable for their actions.”
At the 2016 Ugandan district elections Dr Jablonski and his co-authors and partners conducted a large study of around 16,000 voters. Working with Twaweza, a Ugandan-based organisation that promotes good governance, the research team used mobile phone text messages to inform voters about suspected budget fraud by local government councils.
The results showed that voters who confirmed receiving messages and who learned that suspected fraud was greater than they expected were 6 percentage points less likely to vote for incumbent councillors. Meanwhile, voters who learned that fraud was less than expected were 5 percentage points more likely to vote for incumbent councillors.
In close elections, such margins can be the difference between ejecting a corrupt official or keeping them in office for a further five years.
The team found the messages had no effect in higher profile chair elections. Dr Jablonski says that “this may imply that civil society has less ability to spread credible information in elections where voters are exposed to a lot of information, or where there is a lot of partisan or ethnic voting.”
The project was also part of a larger initiative called “Metaketa I” which sponsored research teams to study the effects of information on voter behavior in Benin, India, Brazil, Burkina Faso, Uganda and Mexico. Most other studies found mixed or inconclusive results, highlighting the challenges in designing effective and scalable civic interventions.
Dr Jablonski views the use of mobile technology as potentially key to the effectiveness of this intervention. He says “Mobile technology allowed us to deliver information directly to citizens in a way that was private, attributable, and tailored to voters’ communities.”
But mobile technology is not without challenges. One of the problems witnessed in developed democracies in recent years is that the ubiquity of mobile phones and personalised political messaging has meant that voters are increasingly ignoring their phones and messages. The risk that voters in countries such as Uganda go the same way when this type of political communication become more familiar was a concern for the researchers.
But the voters who participated did not react in this way; at the end of the study over 80% claimed to have read the messages, and almost all of these found the information valuable.
Dr Jablonski feels that this study highlights the importance of carefully managed relationships alongside technology. He says, “We visited people’s villages individually to explain the project, and we gained their permission to contact them over a period of time. I suspect the trust we built mattered a lot.”
Dr Jablonski sees potential to scale up the research to larger groups of voters, if their research model is replicated, but he warns others who might be inspired by his study against overreliance on technology.
He says: “If somebody just goes into country, sends out a lot of text messages and expects to change people’s minds, they’ll be disappointed.
“You have to build the relationships with the individuals, you have to have a trusted reputation, and you have to provide information that voters believe to be new and valuable.”