Hope City does not yet exist, except as shimmering images of giant beehive towers bathed in a golden light. But Ghana is pressing ahead to build this $10bn (£6.6bn) technology city outside the capital, Accra.

Designed by OBR architects, the city will consist of six towers, including a 75-floor, 270m-tall (885ft) building, the highest in Africa, where 25,000 people will live and 50,000 will work. There have been teething problems, with the site moved to a larger one at Pampram. Residents at the original location, Dunkunaa, claim the project had incurred the displeasure of the gods when the developers refused to acknowledge the chiefs and elders of the area.

That could be the least of Hope City's problems, warns Professor Vanessa Watson of the University of Cape Town. She has flagged up concern not just over the Pampram project, but over other "urban fantasies" involving large-scale reconfigurations of existing cities (such as Kigali in Rwanda, and Addis Ababa in Ethiopia) or the construction of satellite cities (such as Tatu, outside Nairobi in Kenya, and Kigamboni, outside Dar es Salaam in Tanzania). These new city plans are a relatively recent phenomenon, most dating from the past five or six years, with the private sector playing a dominant role in nearly all the projects.

In the paper African urban fantasies: dreams or nightmares?, published by the International Institute for Environment and Development on Monday, Watson casts doubts over whether these grandiose urban projects, driven by local politicians and global investors who see Africa as the "last frontier", will meet the real needs of most people.

"The bulk of the population in sub-Saharan African cities is extremely poor and living in informal settlements," she writes. "Some of these settlements are on well-located urban land that is also attractive to property developers. Attempts to implement these fantasy plans within existing cities will (and is already) having major exclusionary effects on vulnerable low-income groups through evictions and relocations."

She questions the premise on which these big projects are based: rapid urbanisation in Africa and the growth of a middle class on the continent. Watson cites studies arguing that claims of very rapid urbanisation may be overstated and rates may be higher in smaller towns. The Africa Development Bank defines the middle class as those spending $2-20 a day, and the upper middle class as spending $10-20 a day.

"It is difficult to imagine how households with such minimal spending power can afford the luxury apartments portrayed in the fantasy plans (as well as the vehicles needed to move around these new cities), and it may be that prospective property developers are seriously misreading the African market," she writes.

As an example of an architectural white elephant, Watson cites the Chinese-built "ghost towns" outside the Angolan capital, Luanda – cities comprising of tower blocks of apartments selling at $150,000 to $200,000 each. Ambitious urban projects may have worked in Shanghai, China and Ho Chi Minh City, and cities such as Dubai, Shanghai and Singapore are clearly the models for the new African cities.

But Watson doubts whether African governments have the capacity to pull them off. On a mundane level, she wonders how cities, invariably described as smart and sustainable and consisting of tower blocks reliant on lifts and air conditioning, can function in an environment where power cuts occur several times a day.

"Maybe they will be able to, but what nobody can tell me is what will happen to the people living in shack settlements in those cities," Watson says.

Satellite cities are frequently justified on the basis that they are located on empty land. But it is rare, she argues, for larger cities to be empty, and if such land is not within an environmentally protected area, then it is likely to be actively farmed.

"In all kinds of eviction processes, landowners rarely hold land title, and full compensation for land, shelter and livelihoods is unlikely," she adds.

The desire to construct fantasy cities is bound to divert scarce resources from meeting the basic services and housing needs of the much larger poor urban populations. "What seems most likely is that the majority of urban populations will find themselves further disadvantaged and marginalised," writes Watson. "It is access to land by the urban poor (as well as those on the urban periphery and beyond) that is most directly threatened by these processes, and access to land in turn determines to urban services, to livelihoods and citizenship."