Emerging markets (EMs) account for $43 trillion of this, reflecting their greater development needs and faster economic growth, while developed markets (DMs) require $21 trillion (Chart 1).

Building this infrastructure will be crucial for supporting future growth and achieving sustainable development goals. With that, significant amounts of private capital will be required to help fund this investment.

Laying a solid foundation for global growth

As we outlined in our previous research on long-term global growth prospects, different countries and regions face radically different long-term economic futures. Varying population dynamics, stages of development, growth models, political systems, and institutional strength all influence potential economic growth and prosperity.

Physical infrastructure, such as roads and rails, power generation, and utilities, is a keystone within the building blocks of growth. Good infrastructure cuts the cost of doing business, for example, by lowering the cost of producing goods and moving them around the country, which then filters through the vast network of interconnected firms and consumers.

Emerging markets may have the largest needs …

EMs with rapidly growing populations, such as India or Nigeria, and rising urbanization rates generally face the greatest need to expand their infrastructure. Relatively low urbanization rates – particularly in developing Asia – imply that infrastructure spending will increase as workers migrate from rural jobs with low productivity and wages to cities, where prospects are brighter (Chart 2).

… But developed markets also face substantial “gaps”

The accumulation of years of declining public infrastructure spending (Chart 4) also means that many DMs have substantial infrastructure needs, even before considering climate-related expenditure.

Infrastructure in the US was given a “C” rating in its last report card by the American Society of Civil Engineers (ASCE).¹ Many roads, airports, schools, and the energy sector infrastructures were judged to be in poor condition, with “D” ratings. The ASCE flagged that the structural integrity of dams and levees is increasingly affected by extreme weather events, while the number of “high hazard” dams has risen by 20% since 2012.

Power-generation capacity needs to surge

Energy grids across the world need to undergo a massive expansion to accommodate economic growth, the move towards renewable sources of power, the electrification of transportation, and emerging technologies, such as artificial intelligence (AI) data centers.

Much of this demand reflects a combination of structural changes (economic growth, the size of manufacturing sectors, urbanization, and demographics) and the need to shift transport away from fossil fuels and onto the grid. But electricity-generation capacity also needs to expand more than what is implied by demand growth alone.

That’s because the shift to renewable energy requires a greater amount of capacity to be installed compared to thermal power. Simply put, the limited amount of sunshine each day and the unpredictability of wind necessitate a larger capacity to collect and store energy.

Moreover, there is a risk that we are underestimating power needs due to the power-hungry nature of AI. The typical AI data center needs the capacity of around 100 megawatts (MW), equivalent to the annual electricity needs of 100,000 households.² Larger and more power-hungry centers used for more complex models, alongside increasing demand for AI-related services, could see these energy needs rise dramatically.

How can we calculate needs?

Data limitations make estimating precise infrastructure needs far from straightforward, even when we limit ourselves to considering physical infrastructure (such as transport, power, and utilities), rather than broader definitions that could include social infrastructure (such as education, health, and public housing).

Moreover, judging the path for infrastructure is dependent on economic size, which in turn depends on the infrastructure. A 2022 World Bank Review found that each dollar of public infrastructure spending generates $1.50 in additional economic output, suggesting a positive feedback mechanism exists that could push countries onto stronger growth paths, allowing even more investment to take place.³

The World Bank notes that there are relatively few studies to help judge the feedback loop for EMs, but recent work supports the assertion that the effect is stronger. Typically, infrastructure is of poorer quality in EMs (Chart 6), suggesting that improvements can create a bigger boost.

A country with an underdeveloped road and rail transport system may find that addressing these infrastructure shortfalls raises whole-economy productivity, which in turn opens the door to further investment from both the public and private sectors as its economy grows.

Such a dynamic would be most pronounced in EMs but could also operate to a lesser extent in DMs that have underinvested for long periods. However, we must distinguish between what could be and what is likely.

History is littered with examples of EMs failing to emerge, suggesting that budgetary, institutional, and other political and practical constraints often hinder the formation of a virtuous cycle between infrastructure and growth.

Our estimates of infrastructure spending needs are therefore conditioned on our long-term growth forecasts and then combined with other key explanatory variables such as stage of development (GDP per worker), population trends, urbanization rates, and current economic structures (for example, the shares of industry and agriculture).

At a high level, infrastructure needs moderate as income levels rise, and the extent to which countries have already built out their physical infrastructure. However, countries with substantial existing infrastructure must also allocate more funds for maintenance to counteract wear and tear.

How much does the world need to spend?

Investment in the road network makes up the bulk of the global expenditure on transport (around $28 trillion across EMs and DMs), mainly reflecting the seven-million-km expansion of the international road network (+20%), we believe, but also the high maintenance costs of existing roads.

India and China are likely to account for most of the new roads built, while we see some scope for US roads to expand too (+0.4 million kilometers). In contrast, we project little change in the road network in Europe, partly reflecting more muted population and growth trends (Chart 7).

That said, the size of the existing European road network comes with substantial maintenance costs, meaning that it is still one of the largest infrastructure expenditures

worldwide ($1.8 trillion). Rising power needs and the pivot towards renewable energy sources imply more than $27 trillion of expenditure, a notable rise from that of the prior 25 years (Chart 1).

Our models project that global generation capacity needs to rise from 8,000 GW to over 21,000 GW (+165%) by the middle of this century. Less than half of this staggering increase is due to rising demand from global economic growth and structural change. The pivot towards renewables, which require larger infrastructure investment costs upfront to replace an equivalent amount of thermal power capacity, explains around two-thirds of the rise.

China plays a crucial role in the global energy landscape, exemplifying this dynamic. Absent a move towards renewables, we would have expected electricity generation capacity to rise from around 3,000 GW to 5,250 GW by 2050. But the lower "capacity factor" of renewables means we expect China’s generation capacity to rise to over 9,500 GW (Chart 8).

As a result, China’s expenditure on power generation is set to be the largest single infrastructure investment undertaken by any country: $12 trillion is almost 45% of global power generation expenditure and nearly a fifth of total global infrastructure spending over the next quarter century.

Moreover, the risks are skewed towards an even larger and faster expansion of power generation. Leaning into power generation may help China offset drags from real estate and trade tensions with the US. Building up excess capacity may even be part of the country’s long-run strategic aims. Power could be exported to neighboring countries, which would tie them to China politically.

How will future needs be funded?

Both public and private capital play a role in funding infrastructure. The former can be cheaper because governments can borrow at lower rates; however, many governments are feeling the squeeze from higher interest rates, elevated debt levels (Chart 9), and geopolitical pressure to raise defense spending and insulate supply chains.

We expect that private sector involvement will be called on to help fill the gap formed by more reticent governments, bringing efficiency, capital discipline, innovation, and counter to some pre-conceptions, a long-term view removed from election cycles.

¹ "A Comprehensive Assessment of America's Infrastructure." American Society of Civil Engineers, April 2025. https://infrastructurereportcard.org/wp-content/uploads/2025/03/Full-Report-2025-Natl-IRC-WEB.pdf.
² "AI is set to drive surging electricity demand from data centres while offering the potential to transform how the energy sector works." International Energy Agency, April 2025. https://www.iea.org/news/ai-is-set-to-drive-surging-electricity-demand-from-data-centres-while-offering-the-potential-to-transform-how-the-energy-sector-works.
³ "The effectiveness of infrastructure investment as a fiscal stimulus: What we’ve learned." World Bank, February 2022. https://blogs.worldbank.org/en/ppps/effectiveness-infrastructure-investment-fiscal-stimulus-what-weve-learned.

Important information

Projections are offered as opinion and are not reflective of potential performance. Projections are not guaranteed and actual events or results may differ materially.

AA-030625-194402-1