It is said that the first step in overcoming a problem is first admitting its existence. So, here goes: Contemporary public administration is data-challenged.
This would have been an implausible statement to utter, historically. After all, public administrators as individuals know how important data is to public policy formulation and program delivery. Public administration has proved its worth over time with the value of record-keeping, and creating and using data — recording, ordering, sorting and tabulating counts of people, forests, geography, geology, tanks, guns and things like the production of butter.
Indeed, the two great and insatiable needs of the early state, formulated by Yale scholar James C. Scott, were taxation and conscription. Without revenues and the capacity to pay to defend sovereignty, states are not durable. In turn, without public administrators recording, ordering, sorting and tabulating data, the state does not endure.
Historically, public administration has been on the cutting edge of data. Entities often went to various state organs and state registries for data. The public service apparatus of the state knew, even in the state formed explicitly to curb government involvement in the daily affairs of its citizens.
But something dramatic has happened. The administrative state – that part of government that continues regardless of whether elections yield majorities or minorities that are red, blue, orange, green, or purple – is no longer on the cutting edge of data. Yes, the state still knows, but often it only now knows after, while private sector entities know now. Even more powerfully, with predictive analytics, sophisticated private entities increasingly know before.
How can we understand this switch? How can we understand public administration losing its historical position of relative data supremacy? To do that, we need to detour from public administration for a moment and veer into the private-sector economy. What we find gives us important clues to our mystery vis-à-vis data and public administration.
The factors of production
Since Adam Smith, we have understood three core factors of production: land, labour and capital. There are others that have competed to be added to this list. Channeling Peter Drucker, some have argued for “management” – those who direct resources. Others have argued for “entrepreneurs” – those who combine resources in new and innovative ways. But Smith’s formulation has proven remarkably durable for more than two centuries.
If Smith were to return and look at some of the most valuable and dynamic corporations of our era – the digital giants Google, Meta (formerly Facebook), Amazon, Apple, Spotify and others – he would likely be mystified. Yes, he would see some land. Yes, he would see some labour. But nowhere near enough to justify the heady heights – and incredible influence and power – of the digital giants. Finally, he would also see some capital. But remarkably, that capital would largely be a by-product of “production,” and not a driver of production.
Seeing the most valuable and powerful entities on earth during his era, Smith would have seen people – lots and lots of labour. He would have seen land. He would have seen capital in the form of constructed ships, and tools, and extracted then refined natural resources. He would have seen stuff – tangible things that he could touch.
But the contemporary Adam Smith would see negligible amounts of people and land in today’s largest companies. Certainly nothing approaching their value, status or their power. These companies, perhaps most surprisingly of all, “consume” relatively little capital.
So if you are generating enormous profits but not drawing heavily on the “factors of production” …. something makes no sense. What is going on?
Brains? Computers? Digital? Algorithms? Cloud computing?
Yes, yes, yes, yes, yes, and lots more.
But fundamentally, what is going on now is the fourth factor of production.
Data as differentiator
Data has now become the most valuable commodity on earth. Data stocks are more valuable than natural resources. Data is more valuable than manufacturing facilities; more valuable than land; more valuable than labour. Data – the new oil? Oil should be so lucky.
Data is now the differentiator. Data is now the value-add. As computers, software, micro-processing power, storage, cloud computing and algorithms all become (or all trend toward) commodity status, it is the quantity and quality of data that will transform the mediocre into the successful.
Stay in the know with veteran reporter Kathryn May. Sign up for routine and out-of-the-ordinary news about the public service with The Functionary, our new newsletter.
A commodity is an interchangeable and undistinguished part. Where I buy a barrel of oil or a bar of gold or a truck load of gravel or road salt is overwhelmingly just price-contingent. The lowest price wins. To avoid becoming a commodity in data – valued only for how cheaply you can deliver something – you need more and better data than the competition. Increasingly, if you are data-deficient, you will not be competitive or sustainable as an entity.
Put another way, Company A and Company B already compete based on the quantity and the quality of their data. This will also increasingly be true in the coming years for Country A and Country B. Countries have competed forever for oil and gas and timber and nickel. Now they are also adding “quantity and quality of data” to that list of competitions.
Spotify is a data company that deals in music. Netflix is a data company that deals in entertainment. Tesla is a data company on wheels. Google is a data company that deals in information. Amazon is a data company that provides many things – same with Instagram, same with Facebook.
Computing, computation, communication, software, digital distribution – all are, or are rapidly becoming – commodities. Algorithms still have differentiating value, but as advances in artificial intelligence continue, these as well will also invariably trend to commodity status. What really adds value in production increasingly is the quality and quantity of data.
Data and public administration
What does all this have to do with public administration? At first glance, perhaps nothing. But on closer examination, a great deal.
The digital giants became digital giants because they understood – before others – the enormous value of enormous quantities of data. They understood – like the early state understood the power of knowing the quantity and location of trees and people and minerals – that data is power.
As Shoshana Zuboff expertly describes in The Age of Surveillance Capitalism, data becomes the nexus of power. But the power of data in the contemporary age isn’t about counting trees and people, it is rather about the “instrumentalization of behavior for the purposes of modification, prediction, monetization, and control.”
Contemporary public administration, which traces its very heritage back to data, is far less sophisticated in data today than the digital giants. Data is not utilized for public good applications anywhere near the degree to which data is utilized for commercial gain.
Over time, that will harm us all because the public-good realm will have less access to rich data than the private profit realm. Over time, that will make public administration a dinosaur. We need to better understand the power and application of data.
Public administration and real-time actionable data
States often revert to using blunt policy instruments because public administrations do not have the granularity of data – in real time – that is available to the digital giants. When you don’t have real-time actionable data, you estimate. You ask people to apply. You create programs with criteria instead of directly apply funding to public policy objectives.
That worked for a world when real-time actionable data either did not exist or was enormously expensive to actualize. But that is not today’s world. The percentage of the economy migrating online is growing every day, and the online economy has grown much faster than the analog economy in recent years. But something else is happening, too. With the internet of things (IoT), our toasters and our refrigerators and our lightbulbs and our ventilation systems and our water treatment plants and our garage doors and our pacemakers are all migrating online. The enormous oceans of data we have today will, in a few very short years, look like little trickles of water when the IoT begins to take hold in full flight.
Public administration is already behind. Imagine what happens when the volume of data being generated every moment of every day by billions of connected things across the globe increases at an even faster rate.
Does public administration understand the power of data? Do we understand how to use it to serve public policy goals? Do we understand how to regulate it for the public good? Do we have the systems in place to capture data? Do we have the systems in place to safeguard data? Do we have the systems in place to safeguard its use by non-state actors?
These are the many questions facing public administration today. The faster we get the answers, the better public administrators will be able to serve their political decision-makers and their state populations.
Time is not our friend on these questions.