The share of firms that identify access to finance as a major constraint varies widely across economies, but the dataset should not be read as a single-year global ranking. World Bank indicator GFDD.AI.36 reports the percentage of firms identifying access to finance as a major constraint. The dataset contains 148 economies and uses the most recent non-empty observation available for each one. Those observations span 2006 through 2021, so the year attached to one economy can be more than a decade apart from the year attached to another.
That time structure changes how the numbers should be interpreted. Across all 148 latest-available observations, the median is 18.95% and the unweighted mean is 25.27%. Restricting the view to the 51 economies observed in 2019, 2020 or 2021 produces a median of 11.9% and a mean of 16.75%. The lower recent-group figures are not evidence of a worldwide time trend because the set of economies is different. They are a reminder that coverage and observation year are part of the data, not footnotes to be ignored.
Table of Contents
What the indicator measures
The World Bank indicator page describes GFDD.AI.36 as the percentage of firms identifying access to finance as a major constraint. A value of 30% can therefore be read as roughly three in ten firms in that economy’s relevant observation identifying finance access as a major business obstacle. It is an enterprise-reported constraint measure, not a direct measure of bank lending volumes, policy interest rates, the number of bank branches or the total stock of private credit.
A high value signals that access to finance was widely identified as an important obstacle among the firms represented in that observation. It does not, by itself, reveal the mechanism behind the problem. Financing cost, collateral requirements, availability of suitable products, firm size and broader economic conditions may matter, but this one indicator cannot separate those channels. A low value likewise does not prove that every dimension of business finance is easy; firms may face other constraints that rank as more severe.
The center of the 148-observation distribution is 18.95%
For the 148 latest-available observations, the first quartile is about 10.98%, the median is 18.95% and the third quartile is 38.8%. The middle half of the observations therefore lies between roughly 11% and 38.8%. The unweighted mean of 25.27% sits well above the median because a substantial upper tail pushes the average upward.
A simple banding makes that spread concrete. Twenty-six economies are below 10%, 49 are from 10% to below 20%, 37 are from 20% to below 40%, and 36 are at least 40%. Fifteen observations are at 50% or more. These counts describe the latest observation available for each economy, not a common-year snapshot. The 40%-plus group, for example, combines data from several different survey years.
The highest values must be read together with their years
The largest values in the latest-observation dataset are Burkina Faso at 75.0% in 2009, Guinea-Bissau at 71.6% in 2006, Côte d’Ivoire at 69.1% in 2016, Belize at 66.8% in 2010, Sierra Leone at 65.1% in 2017, Dominica at 64.2% in 2010, Mali at 63.5% in 2016 and The Gambia at 63.3% in 2018. Those numbers should not be relabeled as a ranking of which economies currently have the hardest access to finance. The attached years are part of the observation.
The same caution applies at the low end. Hungary is 0.9% in 2019, Eritrea 0.9% in 2009, Panama 1.0% in 2010, Thailand 2.4% in 2016, the Slovak Republic 2.5% in 2019 and China 2.9% in 2012. A comparison that removes the year would mix time differences with cross-country differences and make the result look more precise than it really is.
| Economies | 148 |
|---|---|
| Observation years | 2006–2021 |
| Economies observed in 2021 | 3 |
| Economies observed in 2019–2021 | 51 |
| Median, all latest observations | 18.95% |
| Mean, all latest observations | 25.27% |
| Median, 2019–2021 subset | 11.9% |
| Mean, 2019–2021 subset | 16.75% |
What the 2019–2021 subset looks like
There are 51 observations dated 2019 through 2021. Within this recent subset, the highest values are Lebanon at 47.8% in 2019, Tunisia at 47.4% in 2020, Kosovo at 47.1% in 2019, Ukraine at 46.3% in 2019 and Zambia at 40.1% in 2019. Mongolia follows at 36.8% in 2019, Armenia at 31.6% in 2020, Moldova at 29.4% in 2019, Morocco at 28.4% in 2019 and Türkiye at 26.5% in 2019.
The recent subset has a first quartile of 8.9%, a median of 11.9% and a third quartile of 20.75%. Those numbers are notably lower than the corresponding statistics for all 148 latest observations. However, 97 economies whose latest available row predates 2019 are excluded from this subset. Because the composition changes, the difference cannot be treated as a before-and-after comparison. It describes two different collections of economies.
Only three rows are dated 2021
The three 2021 observations are Austria at 5.1%, Spain at 14.1% and France at 12.5%. That makes the distinction between “latest year in the file” and “year shared by all economies” especially important. The maximum year is 2021, but the dataset does not contain 148 observations from 2021. Most rows come from earlier years.
This is also why a choropleth map labelled simply “2021” would be misleading for this particular extract. A map can visually imply simultaneous measurement even when the underlying rows span many years. The safer presentation is to show the year coverage explicitly and to describe the country values as latest available observations. A true same-year map would require sufficient country coverage for one common year.
Why the measure is still useful
Used carefully, the indicator provides a practical view of how frequently firms report finance access as a serious obstacle. It can help identify economies where financing constraints deserve closer investigation, provide context for business-environment research, and serve as one component in a broader comparison of firm-level constraints. It is especially useful when the observation year is retained and when nearby years are compared instead of collapsing the entire file into one apparently current ranking.
The measure can also be paired with other official indicators, but the concepts should remain distinct. A country can have a deep banking system and still contain firms that report difficulty obtaining finance, while another economy can have lower overall credit depth but a smaller share of surveyed firms calling finance a major obstacle. This indicator captures reported constraint intensity among firms, not the entire structure or performance of the financial system.
Why the median and the observation year matter more than a single global average
The 25.27% mean is easy to quote, but it compresses a skewed distribution and mixed observation years into one number. The median of 18.95% is less sensitive to the highest values, yet it is still a summary of rows from 2006 through 2021. For an individual country comparison, the observation year should be checked first. Comparing economies observed in the same year or within a narrow period generally produces a more interpretable contrast than comparing a 2009 value directly with a 2021 value.
The year distribution is uneven. Thirty-seven observations are from 2019, 19 from 2010, 16 from 2016, 12 from 2009, 11 from 2017, 11 from 2020 and only three from 2021. Older data are not automatically invalid; they remain the latest available observation for those economies in this file. But they should not be presented as though they measured current conditions at the same date.
A practical way to use the dataset
Start by checking both the percentage and the observation year. Next, compare economies observed at similar times. Use the median and quartiles to understand where a value sits within the broader distribution, but keep the mixed-year design visible in the analysis. If a country has an unusually high or low percentage, use the result as a prompt for deeper research rather than as a complete diagnosis.
That approach turns the dataset into more than a ranking table. It shows where firms have reported serious financing constraints, when those observations were recorded, and how the recent-observation group differs from the larger latest-value collection. Keeping those distinctions intact makes the World Bank series more useful for reports, cross-country screening and follow-up analysis without claiming more precision than the data support.
Frequently Asked Questions
Is this a 2021 comparison of 148 economies?
No. The file uses the latest non-empty observation for each economy, and the row-level years range from 2006 to 2021. Only three observations are dated 2021.
What does the 18.95% median mean?
It is the middle value after sorting the 148 latest-available observations. Because those observations come from different years, it should not be described as a 2021 global median.
Does a high percentage prove that the financial system is weak?
No. The indicator measures the share of firms identifying access to finance as a major constraint. It does not directly score financial-system soundness, credit depth, interest rates or bank availability.
What is the safest way to compare economies?
Check the percentage and observation year together, and prefer comparisons among economies observed in the same year or within a narrow time window.
Related Articles
Green Map creates custom-edited map images using open geographic data sources such as geoBoundaries, Natural Earth, OpenStreetMap, and government open data. These maps are edited visual materials, not raw data files, and are provided for education, documents, presentations, and graphic reference.
