Introducing the Worldwide Bureaucracy Indicators

This blog post is part of a series for the 'Bureaucracy Lab', a World Bank initiative to better understand the world's public officials.

Figure 1. How well are public sector doctors paid (relative to their colleagues)? Ratio of public-sector hospital doctor pay to clerical worker pay; by country, presented by region. 
_{Source: Worldwide Bureaucracy Indicators v2}

What proportion of workers around the world are employed by the public sector?  (A lot!  16 percent of total employment, 30 percent of wage employment, and 37 percent of formal sector wage employment.)  How do public and private sector wages compare across and within countries? (Depends on the level of qualifications; the more qualified the lower the wage premium.)  And are women paid more or less than they would be in the private sector?  (Generally more.)  Each of these questions could be the start of a lecture, an academic paper, or a response to a Head of Civil Service thinking about reviewing wage setting policies. But what is the best way of answering them?

Introducing the Worldwide Bureaucracy Indicators' (WWBI)

Until now, the data to do so was limited. The World Bank’s ‘Bureaucracy Lab’ has contributed to filling this gap by developing the Worldwide Bureaucracy Indicators' (WWBI).  The WWBI data set takes a common approach to creating indicators on public and private sector employment and wages, allowing analysts to compare inside the public sector and across the public and private sectors.  To our knowledge, the WWBI is the largest and most comprehensive global dataset on these issues published to date.

The data set presents 192 indicators covering the demographics of the public and private sector workforces, the relative wages between the two sectors, and the public sector wage bill for 202 countries covering the period 2000-2018. The dataset and accompanying codebook are publicly available here and an online dashboard allowing one-click for viewing and exporting is located here.

So now we can return to the questions this blog began with, either country-by-country or taking a cross-national view.    We can say that the public sector employs roughly a third of the world's paid workforce, with Figure 2 outlining how this varies across countries and regions.

Figure 2. Proportion of paid workers employed by the public sector; by country, presented by region. 
_{Source: Worldwide Bureaucracy Indicators v2}

We can take a consistent approach to working out the wages of comparable individuals in the public and private sectors across the world and say that the extra amount that comparable individuals are paid in the public sector rather than the private sector (the ‘public sector wage premium’) is typically positive but heavily mediated by seniority.

Finally, we can see that women are not paid as well as men with comparable characteristics in the public sector, but the penalty is on average lower than in the private sector.  Figure 3 charts the female to male wage ratio for the public (dark blue) and private (light blue) sectors, and typically women fare better in the public sector.  Women are paid an average of 88% of comparable men in the public sector, which is higher than the average of 75% we estimate for the private sector.

Figure 3. Female to male wage ratio in the public and private sectors (using mean); by country. 
_{Source: Worldwide Bureaucracy Indicators v2}

We can then go inside a single government and assess how different government workers are paid relative to their colleagues.  Figure 1 presents the example of doctors versus clerical workers, illustrating the relative wages of doctors employed in public sector hospitals versus their colleagues employed in clerical work.

More on the methodology

Our approach to building the WWBI has been to define a set of uniform country-level indicators based on the Bank’s library of harmonized household surveys. Various teams within the World Bank have expended great effort in collecting household survey data and harmonizing them using a common taxonomy. We then apply a consistent framework to each survey to calculate our indicators.  Our approach is further described in this article in the Public Administration Review [http://dx.doi.org/10.1111/puar.13355] and in the technical note [https://datacatalog.worldbank.org/dataset/worldwide-bureaucracy-indicators] that accompanies the data set. In the service of transparency and reproducibility of the dataset, the code used in the construction of the dataset is archived within the World Bank’s Github repository here [https://github.com/worldbank/Worldwide-Bureaucracy-Indicators].

The result is a micro-founded approach to understanding the labor markets of public administrations and their relationship to private ones. This is the third iteration of the indicators, with version 1 released in December 2018 and 1.1 in June 2020. Version 2 improves the coverage of the dataset across geographies and time and includes a slew of new indicators that present new aspects of the public sector workforce, including its decomposition across industries and occupations of employment.

What can you do now?

The WWBI are an opportunity for policymakers to better understand where their country fits in the landscape of public sector employment and pay and for academics to understand stylized facts of public service compensation. The focus on cross-national data provides a common benchmark for policy makers considering wage setting in their public service and a framework for more detailed analysis.  In providing information on the patterns within public employment, the WWBI enables practitioners and researchers to gain new insights on the public sector across the world. We hope they will be useful for the early slides of your lectures, papers and policy memos.

We hope to continue to improve on the data set with further versions. We continuously engage with academic and practitioner audiences on making the WWBI a stronger product. By releasing the underlying code of our analysis, we invite others to extend our simplified approach suitable for cross-country analysis to include other variables available at the country-level to deepen the analysis for any particular locality.  We would be happy to hear from you on your suggestions as to how we might improve our approach. So, feel free to get in touch.