Brian Klass has shared some interesting speculation on this question at Good Governance Africa. He makes several good points, but I don’t think this is the whole story. From his article:
Only 16% of roads in sub-Saharan Africa are paved—the world’s lowest rate by a wide margin (58% of South Asia’s roads are paved), according to the World Bank.
African governments have not built needed roads or maintained existing ones. This sluggishness runs against strong evidence that financing infrastructure is a valuable long-term investment that creates almost immediate payoffs. Every $1 of public infrastructure spending can contribute up to $0.25 in annual GDP, according to a 2012 World Economic Forum study. This means that savvy investments can pay for themselves in as little as four years.
So why do African governments neglect infrastructure while claiming to search for ways to lift their economies? Two major reasons: constituencies and a penchant for grand projects.
First, “development funding is driven by constituencies,” explained Todd Moss, a senior fellow at the Center for Global Development, a Washington, DC-based think-tank, in a 2013 report on donor aid efficacy. “There’s a strong constituency for health and education funding but there’s no constituency for infrastructure.” It is hard to imagine a TV ad in the US asking for money to pave a road in rural Mozambique the same way it might plead for funds to sponsor a sick child in Somalia. Similarly, when a road is built, statistics cannot back up its impact the way that a public health NGO might showcase the number of vaccinations it has provided. …
Second, government investment in infrastructure—and foreign donor support—is disproportionately skewed towards “big ticket” items rather than smaller projects and ongoing maintenance. Whether it is an African government or a donor-led aid programme, both entities benefit more from grabbing headlines by funding a massive new dam than upgrading a network of roads lost in the hinterland.
While I’m sure both of these things contribute to underinvestment in infrastructure, this can’t possibly explain all of it. One point here is that there should be an obvious domestic constituency pushing for better roads – business owners. Even if most businesses are state-owned, no enterprise benefits from high shipping costs and late deliveries, so you don’t need an independent middle class of small business owners for this prediction to work.
The other point is that construction is a great tool for patronage. Most countries produce the materials that go into roads (cement, gravel, etc.) domestically, since they’re bulky and low-value. Setting up a political ally with a cement firm and a paving contract seems like a really defensible way of promoting local industry and improving infrastructure while also reinforcing one’s patronage network. I can’t find the citation now, but I’ve heard that Japan has proportionately more roads than other countries because construction contracts are such a mainstay of patronage there.
What else do you think might lead to underinvestment in infrastructure?