data analytics hub in the public sector

How government analytics can improve public sector implementation

VoxDevTalk

Published 20.11.24

What do we know about the implementation infrastructure in governments? How can we learn more, and what benefits would improved government analytics have for citizens?

Read “The Government Analytics Handbook: Leveraging Data to Strengthen Public Administration” Edited by Daniel Rogger and Christian Schuster 

While there is a lot of academic research on the design of public policies, less is known about the implementation infrastructure in place within governments. In this episode of VoxDevTalks, Daniel Rogger and Christian Schuster discuss their efforts to use the data that already exists in governments to better understand how they function. 

What is government analytics? 

“We define government analytics in our book as the collection or repurposing of administrative and survey data to better measure and ultimately improve the core public administration” 

One key role of governments is record keeping. While these records used to be paper based, over the last two decades, administrative operations have been digitised in most countries across the world. Therefore, we now have digital records of millions of government transactions. This is a hugely valuable source of data for a new form of analytics. 

Government analytics use this data to understand the administrative machinery of government, that takes inputs into government such as goods, personnel and capital, and converts them into outputs and outcomes for the public. Understanding the processes which occur within governments can help us make better choices about how governments are designing and implementing policies. 

“Over the last 20 years, we've seen a revolution in how deeply we've measured the world.” 

This has helped us appreciate how diverse public administration is, which has huge implications for how we do public policy. 

The importance of understanding how governments function 

Daniel Rogger and Christian Schuster highlight how the microdata now available to researchers can help them dig into specific questions within small organisations and teams, which could potentially have significant aggregate consequences. 

“If you analyse public procurement data across the world, you are analysing 12% of global GDP. If you analyse public payroll data across the world, you are analysing 10% of global GDP, finding margins for improvement there can have very significant global welfare returns.” 

Despite this, there is currently not enough analysis using this government analytics approach. Rogger and Schuster point to a recent survey they conducted which reveals that there are not enough resources being directed towards analytics. This is not just a problem within governments; external researchers also fail to prioritise evidence-based implementation.  

“We often say we want evidence-based policy, but surely we should be saying exactly the same thing, turning the same lens around, and using it to understand how we should best implement.” 

This is one area in which the public sector has fallen behind the private sector. Top technology firms in particular have really understood the value of having a culture of analytics. 

How can data complement the role of public sector management? 

The objective of government analytics should not be seen as a way of ‘measuring away the manager’, but is instead an effort to enrich managerial knowledge and discussions around the use of data in the public sector. 

This does require a culture change towards more proactively wanting to learn from the data, and when it points to shortcomings, accepting the transparency and potential accountability of poor performance. Rogger and Schuster are confident that this culture change can happen for public administration, given how we have quickly come to accept the outcomes of advances in household surveys and randomised control trials in public policy. 

How can governments act on findings on data they collect on public administration? 

Rogger and Schuster’s handbook has a chapter that focuses on exactly this question, by looking at the experience of six governments that run annual surveys of their public servants. High quality information and managerial learning are not sufficient for governments to act on this. Managers also need to have the capacity and motivation to act on results. 

Governments need to invest in data repurposing infrastructure 

Governments typically already have the data collection infrastructure required to analyse public administration, as this is collected by digital systems for government operations that are already collecting data. However, this data is not created for the purpose of analytics, and so this data needs to be repurposed and stored in a way that can enable analysis. Rogger and Schuster propose that governments could create a government analytics unit in the centre of government. It is also important that governments look to attract and retain data scientists, while improving the general stock of knowledge within administrations about the value of data analytics. 

What can we learn from a global survey of public servants? 

Rogger and Schuster have been part of an international research initiative that aims to better understand how public services work by creating high quality, large scale, comparative survey data from public servants around the world. 

“We try to generate that understanding through the lens of public servants... we think that lens is important because it gives us insights into several key determinants of public sector productivity, for instance, the quality of management and leadership.” 

This data is particularly useful for cross-checking the stereotypes about public servants with the realities of their work. For example, many assume that they do not have performance incentives. However, in most countries that were surveyed, public servants say that their performance plays an important part in promotion decisions. 

Further insights can be gleaned on the strengths and weaknesses of public administrations. For example, evidence shows that the majority of public servants are satisfied with their jobs, but not with their pay. 

These survey results also highlight the incredible diversity in the experiences of public servants. Typically, there is more divergence between experiences in different government organisations in the same country than between the governments across countries.