What recent trends do we see in countries’ PFM performance?


This page in:

In 2017, we took a close look at trends in Public Financial Management (PFM) reforms by examining Public Expenditure and Financial Accountability (PEFA) and Country Policy And Institutional Assessment (CPIA) data. This was to understand patterns and trends, and specifically, to explore if and how closely PFM performance is linked to external factors such as country size, income levels, or natural resource dependency as well as political factors, and the level of domestic revenue collected. In addition, the study covered five cases in depth to re-trace how PFM reforms were initiated and implemented. 

At that time, we found four key patterns in the data: 

  • First, aggregate changes over time have been limited, especially outside of Sub-Saharan Africa and Eastern Europe and Central Asia – with PEFA data suggesting somewhat greater improvements in PFM performance than CPIA 13 data (which assesses ‘quality of budgetary and financial management’). 
  • Second, wealthier, more populated, and non-resource rich countries performed better than those with the opposite characteristics. 
  • Third, a significant and encouraging relationship, countries with lower starting points saw the greatest degree of improvements. 
  • Fourth, across the different ‘buckets’ of the PEFA framework, countries tend to perform best in policy- based budgeting and worst in external audit and scrutiny.   

Three years later, it is time to revisit the findings – and to reflect on what the PFM trends tell us for a world experiencing major shocks. Do findings change as additional years of observation become available or do trends persist? What new information can we learn about PFM reform efforts made so far, and are there insights that might be useful in facing current challenges and seeking to move further ahead?  

Repeating the analysis done in 2017 expands the number of observations from around 250 to 290 assessments, adding both additional countries and PEFA rounds in the same country (between 14 and 19 national level PEFA assessments have been completed annually in recent years). The results show that the two data sources for assessing PFM performance continue to tell a different story. Comparing average CPIA data by region over time, there are few positive changes. But PEFA averages continue to improve, especially for Sub-Saharan Africa, closing in on the global average equivalent to ‘C+’. Eastern Europe and Central Asia remains the only region that reaches a ‘B’ average on the PEFA scale (see Fritz et al. 2014 and 2017). Among the 20 countries showing the greatest improvement between their initial and the most recent PEFA assessment, 7 are in Africa, and 6 are in ECA. 

Looking at cross-country relationships that explain differences in PEFA scores, the patterns observed earlier have slightly strengthened with a few additional findings: countries with higher income levels, higher economic growth, lower resource dependence, and those that are not small island developing countries do better on average (Table 1). Having higher tax revenue and greater aid dependency do not show a significant relationship with PFM performance either in a cross-country perspective or by looking at differences over time. Cross-country comparisons also continue to show positive relationships between greater political stability, a more democratic regime, and having programmatic parties and PFM performance. 

Average PEFA Scores and Country Characteristics
Note: (a) the Y variable, ‘PEFA score’ is the average of 21 indicators as per the 2011 PEFA framework (PI-5 to PI-12 and PI-16 to PI-28); (b) we exclude PI-1 through PI-4, which measures PFM outcomes, indicators PI-13 to PI-15, which cover revenue administration, and D1 to D3, which are donor related indicators; (c) 13 observations were dropped, and (d) high-income countries were excluded. Standard errors in parentheses, clustered at the country level. Observations weighted inversely to number of PEFA assessments. No influential observations are dropped.
*** p<0.01, ** p<0.05, * p<0.1

Considering dynamics over time (‘first differences’) shows that improving PFM performance is correlated with GDP growth, with the effect strongest among countries starting from lower initial GDP per capita. We also again find that improvements in performance are largest in countries with a weaker initial PFM performance. Given that initial levels of GDP and PFM performance are highly correlated, what helps to increase PFM performance remains difficult to disentangle.  

Next week, we will publish part two which will dive into what these trends mean for a COVID-19 impacted world.


Verena Fritz

Senior Public Sector Specialist, Governance Global Practice, the World Bank

Join the Conversation

Ngemera Ndibalema
October 22, 2020

I am particularly interested in improving PFM at sub-national level governments.