Satellite Data and AI Reveal Detailed Map of Jobs and Employment Patterns in Ghana

Researchers from the International Labour Organization and Nanjing University of Posts and Telecommunications used satellite data and machine learning to create detailed maps showing where jobs are located across Ghana. The study reveals strong regional differences in employment, with jobs concentrated in southern urban areas while northern regions have far fewer opportunities.

A new study by the International Labour Organization (ILO), together with researchers from the School of Internet of Things at Nanjing University of Posts and Telecommunications and the Smart Health Big Data Analysis and Location Services Engineering Lab of Jiangsu Province, is offering a new way to understand Ghana’s labour market. By combining satellite imagery, geographic data and machine learning, the researchers created detailed maps showing where jobs are located across the country. These maps reveal labour market patterns at a much finer scale than traditional statistics, helping policymakers see how employment opportunities vary between cities, towns and rural communities.

Labour data in many developing countries is usually available only at national or district levels. While useful, such data often hides important local differences. Economic indicators like population or GDP have long been mapped using satellite data, but labour indicators such as employment, unemployment, skills and sectoral activity have rarely been studied with similar detail. The new research addresses this gap by breaking down national labour statistics into small grid cells about half a kilometre wide, offering a clearer picture of where people work and what types of jobs exist.

Ghana’s Labour Market in Focus

Ghana is home to about 30.8 million people, with nearly 20 million of working age. Around 9.6 million people are employed, giving the country an overall employment rate of about 48 percent. However, the national average hides big differences among different groups.

Young people aged 15 to 24 face more challenges finding work, with employment rates of only about 23 percent. Prime-age workers between 25 and 54 have much higher employment levels, while older workers fall somewhere in between. There is also a gender gap, with men generally having higher employment rates than women.

The structure of Ghana’s labour market reflects conditions common in many developing economies. Self-employment is widespread and accounts for nearly two-thirds of all jobs. Education levels also shape employment opportunities, as many workers have only basic schooling. Sectorally, services now employ the largest share of workers, followed by agriculture. Manufacturing and industry employ a smaller but growing portion of the workforce.

How Satellites and AI Helped Build the Maps

To create the maps, researchers used information from the 2021 Ghana Housing and Population Census. They combined this data with 64 different types of geographic and satellite-based information. These included land cover, rainfall, vegetation health, road networks, building density, and nighttime light intensity, which is often used as a measure of economic activity.

A machine-learning method called a random forest model was then used to analyze the relationship between these factors and employment levels. The model was trained using district-level employment statistics and then applied to estimate employment in each small grid cell across the country.

This approach allowed the researchers to estimate 17 different labour market indicators, including employment by age group, gender, skills, sector, and employment status. It also included measures of unemployment and the number of young people not in employment, education or training.

Clear Regional Differences in Job Opportunities

The maps show that employment is heavily concentrated in southern Ghana. The urban corridor between Accra and Kumasi stands out as the country’s main employment hub. Coastal areas and regions along major roads also show higher concentrations of jobs.

In contrast, northern Ghana has much lower employment density. Economic activity there is mainly centred around a few urban centres such as Tamale, Wa, Bolgatanga and Bawku. These patterns highlight the big regional differences in economic opportunities within the country.

Urban areas also dominate in terms of job quality. Salaried and higher-skilled jobs are mostly found in large cities where government services, offices and industries are located. Rural areas depend more on farming and small self-employed businesses. Manufacturing jobs, while limited nationally, appear in small clusters around certain urban areas, suggesting the presence of local industrial centres.

Why These Maps Matter for Policy

When employment is measured relative to the working-age population, the maps show an even more complex picture. Local employment rates vary widely across Ghana, ranging from very low levels in some places to nearly full employment in others. Overall, southern regions tend to have higher employment rates than the north, although some rural areas also show strong participation in farming and informal work.

The study also identifies factors that strongly influence employment patterns. Areas with more buildings, brighter nighttime lights and better road networks tend to have more jobs because these features reflect economic activity and easier movement of goods and people. Agricultural employment is closely linked to environmental conditions such as rainfall and vegetation health.

For policymakers, these detailed maps could become a powerful planning tool. They can help identify regions with fewer job opportunities, highlight areas where infrastructure could support economic growth, and guide targeted job creation programmes.

More broadly, the research shows how combining census data with satellite imagery and machine learning can transform labour market analysis. Countries that lack detailed economic surveys may still be able to generate high-quality employment insights using similar methods.

For Ghana, this new approach offers a clearer understanding of where jobs are located and how labour markets function across the country. With better information, governments and development partners can design policies that address regional inequalities and support more inclusive economic growth.