What if you could truly compare the size of economies, the standard of living, or the cost of a meal across countries—without the distortions caused by currency fluctuations or wildly different local prices? That’s the promise of the International Comparison Program (ICP), a global statistical initiative that underpins how we understand economic power and poverty around the world. But how does it actually work, and why do policymakers and researchers rely so heavily on its results?
Short answer: The International Comparison Program (ICP) is a global partnership led by the World Bank under the United Nations Statistical Commission, established to produce Purchasing Power Parities (PPPs) and related price level indexes. Its purpose is to enable robust, comparable economic analysis across countries by adjusting for differences in price levels, so that GDP and other macroeconomic aggregates can be meaningfully compared in “real” terms—not just at market exchange rates. The methodology involves collecting detailed price and expenditure data for a common basket of goods and services, validating and harmonizing these data across up to 176 participating economies, and using standardized international accounting frameworks to compute PPPs at multiple levels of economic aggregation.
Let’s unpack what this means, why it matters, and how the ICP carries out this ambitious mission.
Why the ICP Exists: The Challenge of Comparing Economies
Comparing a country’s gross domestic product (GDP) or living standards using market exchange rates alone is misleading. Exchange rates can swing due to financial markets, government policies, or speculation, and don’t reflect what goods and services actually cost in each country. For instance, a haircut or a loaf of bread might be much cheaper in Bangladesh than in France, even after converting currencies. This makes it impossible to answer basic questions like: Which country’s people can buy more with their incomes? Where is poverty deepest, once you account for local prices?
The ICP solves this by producing “purchasing power parities” (PPPs)—conversion rates that equalize the purchasing power of different currencies for a fixed basket of goods and services. According to worldbank.org, PPPs are both “currency conversion factors and spatial price indexes” that eliminate price level differences between economies, allowing GDP and other measures to be expressed in real, comparable terms. As cepal.org notes, this is essential for tracking progress on global goals like the Sustainable Development Goals (SDGs), which require “comparable price and volume measures of national accounts aggregates across economies of the world.”
What Does the ICP Produce?
The core outputs are PPPs and price level indexes (PLIs). PPPs convert national GDP (and its components) to a common currency, adjusting for price level differences. The resulting “real” GDPs (and per capita figures) can be used to compare economic size, living standards, poverty rates, and more—across countries or regions, and over time. Price level indexes show how a country’s general price level compares to a reference country or the world average.
For example, the 2021 ICP cycle—the latest as of May 2024—produced PPPs and PLIs for 176 economies, with regional and global reports covering areas from Asia and the Pacific to Africa and Latin America (adb.org, worldbank.org, cepal.org).
How the ICP Works: The Methodology
The ICP’s methodology is intricate and rigorous, designed to ensure that results are accurate, comparable, and resistant to political influence. Here’s how it unfolds:
International Framework and Governance
The ICP is coordinated globally by the World Bank, under the oversight of the United Nations Statistical Commission (UNSC), which designated it as a permanent element of the global statistical program in 2016 (worldbank.org). Regional agencies—like the Asian Development Bank (ADB) for Asia-Pacific or the Economic Commission for Latin America and the Caribbean (ECLAC) for Latin America—run the program at the regional level, working with national statistical offices.
A multi-layered governance structure, including the ICP Governing Board and Technical Advisory Group, ensures that “estimates of PPPs and ICP results are calculated independently based on common, sound, and transparent methodology” (worldbank.org). This robust oversight is crucial to maintaining the integrity and utility of the data.
The Building Blocks: The System of National Accounts
At its core, the ICP relies on the internationally agreed System of National Accounts (SNA)—specifically the 2008 version for recent cycles—to define GDP and its components. SNA sets out strict rules for measuring economic activity, ensuring that all countries are adding up GDP in the same way (icp.adb.org, worldbank.org, adb.org).
GDP is measured from the expenditure side, broken down into final consumption (by households, government, and non-profits), gross capital formation (investment), and net exports. These aggregates are further subdivided: the current ICP framework uses six main aggregates, 28 expenditure categories, 63 groups, 126 classes, and, crucially, 155 “basic headings.” These basic headings are “the lowest level of aggregation of items in the GDP breakdown for which expenditure data are available” (icp.adb.org). For example, “rice,” “footwear,” or “post-secondary education” might each be a basic heading.
Collecting Price and Expenditure Data
Each participating country collects national annual average prices for a carefully selected set of goods and services, chosen from a common basket that is representative and comparable across economies. The selection aims to reflect what people actually buy and to ensure that items are as similar as possible—a challenge, given the diversity of products and services worldwide.
Alongside prices, countries provide detailed expenditure data in their local currency, broken down to the 155 basic headings. This disaggregation is vital: it allows the ICP to weight prices appropriately when aggregating up to the GDP level, ensuring that each item’s importance in the national economy is reflected in the calculations.
Price and expenditure data are typically collected for a reference year—such as 2011, 2017, or 2021. The “reference year” ensures that comparisons are made for the same period, adjusting for inflation or major economic shocks (as occurred when the 2020 cycle was delayed to 2021 due to COVID-19 disruptions; worldbank.org, cepal.org).
Rigorous Data Validation
To ensure quality and comparability, the ICP employs a three-stage validation process. First, countries perform intra-country checks—ensuring completeness, SNA compliance, and plausibility of their own data. Next, regional agencies conduct inter-country validation, comparing data across countries in the same region to spot inconsistencies or outliers. Finally, the global office and technical experts perform inter-regional validation, integrating and harmonizing data across all regions for the final global comparison (icp.adb.org, worldbank.org).
For example, the Asian Development Bank reports using a specialized GDP Price Collection Tool with “built-in validation checks,” ensuring “additivity from the basic heading level to higher levels of aggregation; completeness of entries for the 155 basic headings; completeness of entries for the years 2017 to 2021; and presence of mandatory variables” (icp.adb.org). These checks are repeated at the regional and global levels, making the process robust against errors or manipulation.
Calculating PPPs and Real Expenditures
Once validated, price and expenditure data are fed into the ICP’s complex statistical machinery to calculate PPPs at the basic heading level. These are then aggregated, using expenditure weights, up to higher levels—categories, aggregates, and eventually GDP. The result: a PPP for each country’s GDP, which tells you how much local currency is needed to buy the same basket of goods and services as one unit of the reference currency (such as the US dollar, or, in Asia-Pacific, the Hong Kong dollar).
These PPPs are then used to convert national GDP and its components into “real” expenditures, directly comparable across countries. The ICP also produces price level indexes: ratios of PPPs to exchange rates, indicating whether a country is relatively cheap or expensive compared to the reference economy (adb.org, worldbank.org).
Key Properties: Transitivity and Base Economy Invariance
Two properties are essential for reliable international comparisons: transitivity and base economy invariance. Transitivity means that the direct PPP between any two economies should match the indirect PPP calculated via a third economy—ensuring consistency. Base economy invariance means that the relative positions of economies are unaffected by the choice of reference currency or country (adb.org).
Coverage and Impact
The reach of the ICP is vast: the 2021 cycle covered 176 economies, with regional breakdowns including 21 in Asia and the Pacific, 36 in Latin America and the Caribbean, and 52 in Africa (adb.org, worldbank.org, cepal.org). The program has evolved from intermittent rounds in the 1970s to a regular three-year cycle, with increasing country participation and growing methodological sophistication (worldbank.org, cepal.org).
The results are indispensable for global policy and research. International organizations use PPPs to formulate and monitor the SDGs, set global poverty lines (such as the “$2.15 a day” international poverty line), and compare indicators like income inequality, health, and education expenditure. National governments use ICP data for benchmarking, planning, and policy analysis. As cepal.org notes, the results “enable international comparisons of gross domestic product and its components in real terms, expressed in the same monetary unit and adjusted for the different price levels in the economies.”
Real-World Examples and Applications
The ICP’s influence is everywhere. For example, the World Bank’s estimate of global extreme poverty relies on PPP-adjusted thresholds, allowing meaningful comparisons of poverty rates from Nigeria to India to Brazil. The Asian Development Bank publishes regional PPPs and real GDP figures for 21 economies, underpinning analyses of “living standards and price levels in Asia and the Pacific within the context of the global economy” (adb.org).
Regional organizations like ECLAC in Latin America not only coordinate data collection but also help harmonize statistical methods and improve national statistical capacities, with regional collaboration fostering “the necessary harmonization of methods and procedures and helping to broaden the coverage and quality of these statistics” (cepal.org).
Challenges and Evolution
The ICP faces constant challenges: collecting truly comparable data across such a diverse set of countries is a monumental task, and even “basic headings can cover a broader range of goods or services than is theoretically desirable” due to practical constraints (icp.adb.org, worldbank.org). The COVID-19 pandemic in 2020 forced a postponement of the planned ICP cycle, highlighting the vulnerability of global data efforts but also the resilience of the program’s infrastructure (worldbank.org, cepal.org).
To address data gaps and improve cost-effectiveness, researchers have explored “reduced information” approaches to PPP estimation, especially for countries with limited resources (adb.org). The ICP’s methodology and outputs are constantly reviewed and refined by international expert groups.
Conclusion
In sum, the International Comparison Program is the backbone of global economic comparison, transforming disparate national data into internationally comparable measures that shape everything from poverty reduction efforts to cross-country investment decisions. Its sophisticated methodology, grounded in rigorous international standards and robust governance, ensures that “comparing apples with apples”—to borrow a phrase from worldbank.org—becomes possible, even across the world’s most different economies. With each cycle, the ICP brings us closer to understanding not just how big countries’ economies are, but how far their money really goes.