Global Projects, Real-World Impact

What do projects actually do to economies and to the people who live in them, and what happens when the decisions behind them are poor? The evidence gives a clear answer. Well-chosen projects, delivered efficiently, raise output for years after the money is spent, connect firms and households to markets and services, and support employment well beyond the construction site. Poorly chosen or poorly delivered projects forfeit much of that value, and some leave economies worse off than if nothing had been built.

The International Monetary Fund’s analysis of advanced economies found that an unanticipated increase in public investment of 1 percentage point of GDP raised output by about 0.4% in the same year and 1.5% four years later (IMF, World Economic Outlook, October 2014). Separate IMF work found that around 30% of the potential gains from public investment are lost to inefficiencies (IMF, Making Public Investment More Efficient, 2015). Between those two findings sits the argument of this article: the impact of a project depends less on the amount spent than on the quality of the decision and of the delivery system behind it.

That matters to anyone who approves, funds or delivers projects. Impact does not simply arrive on completion. It is designed in at the decision, protected or lost in delivery, and realised, or not, over decades of operation.

In brief

• Public investment raises output in both the short and the long term. In IMF estimates for advanced economies, an increase of 1 percentage point of GDP raised output by about 0.4% in the first year and 1.5% after four years.
• The effect depends on efficiency: about 2.6% after four years in high-efficiency countries against about 0.7% in low-efficiency ones. The IMF estimates that around 30% of the potential gains from public investment are lost to inefficiencies.
• Connectivity projects shape trade and inclusion. Shipping moves over 80% of world merchandise trade, and 2.2 billion people remained offline in 2025.
• Unreliable infrastructure is expensive: the World Bank estimates that disruptions cost households and firms in low- and middle-income countries between US$391 billion and US$647 billion a year.
• Poor decisions have measurable consequences, from cost overruns and demand shortfalls to delayed revenues. Impact has to be planned at the decision and proven in operation.

Why impact is the real test of a project

Projects are the mechanism through which capital becomes capacity: the power station, the port, the railway, the fibre network, the factory. But capacity is not impact. Impact is the change an asset makes to what an economy can produce, how easily goods, people and data move, how reliably services reach households and firms, and who gains or bears the costs.

The economic case for projects is often misread in both directions. Advocates treat spending as if it were value; critics treat cost overruns as if they settled the question of worth. The evidence supports neither shortcut. Output effects are real but conditional. Failures are common but not universal. And the same project can disappoint in delivery and still change the economy it serves.

The macroeconomic evidence: output, debt and the efficiency condition

Influential evidence on the output effects of public investment comes from Chapter 3 of the IMF’s October 2014 World Economic Outlook. Using data for 17 advanced economies over 1985–2013, IMF staff found that public investment shocks had statistically significant and long-lasting effects on output, implying short- and medium-term spending multipliers of about 0.4 and 1.4. The chapter concluded that debt-financed projects could have large output effects without increasing the debt-to-GDP ratio, provided clearly identified infrastructure needs were met through efficient investment (IMF, World Economic Outlook, October 2014).

Two conditions shaped the result. The first was economic slack. During periods of low growth, output rose by about 1.5% in the same year and 3% in the medium term, while in periods of high growth the long-term effect was not statistically different from zero. The second was efficiency, and the difference it made was large.

The table shows the IMF’s estimated effect of an unanticipated increase in public investment of 1 percentage point of GDP (IMF, World Economic Outlook, October 2014).

The IMF’s 2015 follow-up quantified the efficiency problem. Comparing the value of public capital with measures of infrastructure coverage and quality across countries, it found that around 30% of the potential gains from public investment are lost to inefficiencies in the investment process. On its preferred indicator, the average gap was 40% in low-income developing countries, 27% in emerging markets and 13% in advanced economies. The most efficient public investors obtained twice the growth impact of the least efficient, and the IMF estimated that stronger public investment management institutions could close up to two-thirds of the efficiency gap (IMF, Making Public Investment More Efficient, 2015).

Our reading is that these findings reframe the familiar debate about whether to invest more. On the IMF’s own estimates, the difference between an efficient and an inefficient investment system is of the same order as the average effect of the investment itself. A government or company that improves how it chooses, appraises and manages projects may gain as much as one that simply spends more.

Employment and economic activity

Projects create work directly, in design, construction and supply chains, and indirectly, through the activity that their spending and their services support. The IMF’s October 2020 Fiscal Monitor, drawing on a cross-country data set and a sample of 400,000 firms, estimated that in periods of high uncertainty the fiscal multiplier of public investment in advanced and emerging market economies peaks at over 2 within two years. It estimated that increasing public investment by 1% of GDP in these economies would create 7 million jobs directly, and between 20 million and 33 million jobs overall once indirect macroeconomic effects are included. It also found that crowding-in of private investment was particularly strong in communications, transport, construction and manufacturing (IMF, Fiscal Monitor, October 2020).

Two cautions apply. Construction employment is temporary by nature; the lasting employment effect comes from what the finished asset enables, whether an industrial cluster around a port or new services around a transit line. And the IMF’s estimates were made for the recovery from the pandemic, when slack in the economy was high. Where skilled labour and materials are already scarce, as our analysis of the next decade of global projects sets out, our reading is that additional spending is more likely to bid up costs than to add jobs.

Industrial capacity and long-term competitiveness

Industrial capacity is the accumulated stock of past projects: plants, mines, refineries, grids and logistics that allow an economy to make and move things. Its value shows most clearly under stress. The IEA estimates that investment over the past decade in renewables, nuclear, electrification and efficiency in five major fuel-importing regions (China, the European Union, Japan and Korea, Southeast Asia and India) avoided around US$260 billion in fossil-fuel import costs in 2025 (IEA, World Energy Investment 2026). The IEA expects the benefit to be considerably larger in 2026, a year in which conflict in the Middle East has disrupted energy markets. Those savings are the product of many project decisions taken years earlier.

Competitiveness compounds in the same way. If efficient investors obtain markedly more growth from each unit of capital, as the IMF’s work indicates, then over successive investment cycles differences in project capability become differences in national capacity. The same logic applies inside companies. A firm that consistently brings plants, systems and facilities into service on time and at the expected performance can expand, respond to demand and enter markets faster than competitors that cannot.

Connectivity: trade, ports and digital networks

Projects that move goods have consequences far beyond the places where they are built. Global shipping moves over 80% of the world’s merchandise trade, according to UNCTAD’s Review of Maritime Transport 2025. The latest edition shows what happens when that system is disrupted. Seaborne trade grew 2.2% in 2024, but long-distance rerouting caused by geopolitical tensions pushed ton-miles up by nearly 6%, and UNCTAD forecast growth of only 0.5% for 2025. At the report’s launch, UNCTAD described how rerouting around the Cape of Good Hope has meant higher transport costs, delays and emissions, and a reshuffling of shipping networks affecting ports, including in Africa. It identified small island developing states, least developed countries and net food-importing nations as the most vulnerable, because higher freight costs quickly translate into more expensive imports and food insecurity (UNCTAD, Review of Maritime Transport 2025).

Ports, canals and corridors are therefore economic infrastructure whose performance is felt across entire trading systems. In our view, decisions on port capacity, alternative routes and digital trade systems are competitiveness decisions as much as transport decisions.

Digital networks are now as basic to economic participation as roads. The International Telecommunication Union (ITU) estimates that about 6 billion people, around three-quarters of the world’s population, used the internet in 2025, while 2.2 billion remained offline, 96% of them in low- and middle-income countries. The divides are structural: 94% of people in high-income countries are online against 23% in low-income countries, and 85% in urban areas against 58% in rural areas. 5G networks cover an estimated 55% of the world’s population, but 84% of people in high-income countries and only 4% in low-income countries (ITU, Facts and Figures 2025). Econometric modelling for the ITU across 139 countries estimated that a 1% increase in mobile broadband penetration yields an increase of 0.15% in GDP (ITU, The economic contribution of broadband, digitization and ICT regulation, 2018). Estimates of this kind carry wide uncertainty, but the direction is consistent: where networks are not built, economic participation is constrained. The link between network investment and enterprise change is one reason transformation and technology investment increasingly sits on the same agenda as physical infrastructure.

Reliability: the cost of infrastructure that fails

Infrastructure affects economies not only when it is built but on every day that it works or fails. The World Bank’s Lifelines report estimated that infrastructure disruptions impose costs of between US$391 billion and US$647 billion a year on households and firms in low- and middle-income countries. Using data on about 143,000 firms in 137 of those countries, it found capacity-utilisation losses from power, water and transport disruptions of US$151 billion a year, sales losses from electricity outages of US$82 billion and extra costs of self-generating electricity of US$65 billion, which the authors describe as lower-bound estimates. Natural hazards typically explain 10% to 70% of disruptions, depending on sector and region; other causes include poor maintenance, mismanagement and underfunding (World Bank, Lifelines, 2019).

The effects reach households as well as firms. The report cites research from South Asia finding that long power outages are associated with lower income per head and lower female labour-force participation. Productivity, in other words, is lost not only where infrastructure is missing but where it is unreliable.

The implication for project leaders is direct. Impact depends on operation and maintenance, not only on construction. An asset handed over without the budget, skills and systems to maintain it will deliver less than its business case promised, a theme developed in our analysis of turning projects into productive assets. The investment case for maintenance and resilience is set out in our analysis of infrastructure investment.

When decisions go wrong

Research on megaprojects shows that cost overruns are the norm, that benefit shortfalls are common, and that the projects that look best on paper are often those that perform worst (Flyvbjerg, Project Management Journal, 2014). The mechanisms, and the front-end disciplines that counter them, are set out in why capital projects fail before construction begins. The concern here is what those failures cost beyond the project budget.

The first cost is the one the IMF measures: output that efficient investment would have produced and inefficient investment did not. The second is opportunity cost. Capital, skilled people and political attention committed to the wrong project are unavailable for the right one, and Flyvbjerg concludes that projects often turn out to bring a net loss to the economy rather than a gain. The third is time. A delayed asset postpones every benefit that depends on it, from fare revenue and tax receipts to the homes, jobs and private investment that follow new capacity. The fourth falls on people outside the project: users who wait longer for services, residents who live with prolonged construction, and taxpayers who fund the gap.

One project, two verdicts: the Elizabeth line

London’s Crossrail programme shows how delivery failure and economic value can coexist. The UK National Audit Office (NAO) reported in 2019 that the programme had missed its planned December 2018 opening of the central section and that Transport for London’s 2018 business plan assumed £600 million of revenue losses between 2019-20 and 2023-24 as a result of the delays. The NAO concluded that features of the way the programme had been delivered had driven unnecessary cost (NAO, Completing Crossrail, 2019). In 2021 it found that Transport for London did not yet have a plan for maximising the benefits set out in the business case (NAO, Crossrail – a progress update, 2021).

The line opened to passengers in May 2022. Two years later, Transport for London reported more than 350 million journeys since opening, including 210 million in 2023/24, with demand already above its post-pandemic expectations for the end of the decade. It reported that the line had directly affected the development of 55,000 new homes and estimated that it had added £42 billion to the UK economy; that figure is the operator’s own estimate rather than an independent evaluation (Transport for London, 2024).

The lesson is not that overruns do not matter. Crossrail’s delays imposed real costs on public finances and on the transport authority’s revenues, and postponed the railway’s benefits by more than three years. The lesson is that impact has to be judged over the life of the asset and managed as deliberately as cost and schedule, which is the case for measuring capital project performance as cost, time and value.

What this means for senior executives

Six practical implications follow for leaders who approve, fund and deliver projects.

1. Define impact before approval. State the economic change a project is meant to produce, whether capacity, reliability, access or productivity, and how it will be measured in operation, not only at handover.
2. Invest in the system, not only in projects. The IMF evidence points to the institutions that plan, appraise, select and implement investment as a major source of efficiency gains. Inside organisations, the equivalent is disciplined project portfolio management and investment governance.
3. Budget for maintenance and reliability from the start. Unreliable service imposes costs on users that rarely appear in a project’s own accounts; deferred maintenance is a decision to accept lower impact later.
4. Make the cost of delay visible. Delay has an economic price in forgone revenue and postponed benefits. It belongs in investment decisions and in the reporting of project controls.
5. Ask who gains and who bears the cost. Connectivity gaps between urban and rural areas, and between rich and poor countries, show that impact is unevenly distributed. That distribution shapes public consent and, over time, the value an asset can deliver.
6. Treat benefits realisation as work in its own right. The NAO’s Crossrail finding makes the point: benefits need an owner, a plan and a budget, just as construction does.

Impact is shared across functions

The economic impact of projects is not the concern of one function. Portfolio leaders decide which projects compete for scarce capacity. Leaders of capital projects and CAPEX programmes determine whether assets arrive on time and perform as designed. Funding and governance bodies decide whether money follows evidence of value. Delivery and risk functions protect the schedule on which benefits depend. Technology and transformation leaders now shape connectivity itself. Results and value is where all of this is finally tested.

Decide, fund, deliver, prove: where impact is won or lost

Seen through the full life of a project in the project economy, impact is determined at every stage.

• Decide. The efficiency evidence suggests that a large share of value is won or lost in selection and appraisal, before any money is spent.
• Fund. Investment directed at clearly identified needs, and delivered efficiently, can raise output enough to avoid adding to debt ratios; inefficient investment cannot.
• Deliver. Overruns and delays consume capacity and postpone benefits, as Crossrail’s revenue losses illustrate.
• Prove. Impact must be measured in operation, over years, and reported with the same discipline as cost.

This is the sequence around which Project Economy Forum organises its work, and its Dubai edition on 27–28 January 2027 follows one project along it, from first decision to final proof.

Conclusion: how well, not only how much

The most expensive project is rarely the one that overruns. It is the one that should not have been built, or the one that is built and never allowed to deliver its purpose. The evidence reviewed here suggests that the question facing governments and companies is seldom simply whether to invest. It is whether their systems for choosing, delivering and operating projects are good enough to turn investment into impact. On the IMF’s estimates, that capability can matter as much as the size of the investment itself, which makes it one of the most consequential economic choices any organisation controls.

Sources

• International Monetary Fund, World Economic Outlook, October 2014, Chapter 3, “Is It Time for an Infrastructure Push? The Macroeconomic Effects of Public Investment” (2014). IMF
• International Monetary Fund, Making Public Investment More Efficient (2015). IMF
• International Monetary Fund, Fiscal Monitor: Policies for the Recovery, executive summary (October 2020). IMF
• International Energy Agency, World Energy Investment 2026 (2026). IEA
• UN Trade and Development, Review of Maritime Transport 2025: Staying the course in turbulent waters, launch briefing (2025). UN Geneva
• International Telecommunication Union, Facts and Figures 2025 (2025). ITU
• International Telecommunication Union, The economic contribution of broadband, digitization and ICT regulation (2018). ITU
• World Bank, Lifelines: The Resilient Infrastructure Opportunity (2019). World Bank
• Project Management Journal, Bent Flyvbjerg, What You Should Know about Megaprojects and Why: An Overview (2014). arXiv
• UK National Audit Office, Completing Crossrail (2019). NAO
• UK National Audit Office, Crossrail – a progress update (2021). NAO
• Transport for London, The Elizabeth line continues to transform travel in London on its two-year anniversary (2024). TfL