25/09/2018, 15:54



Demystification-of-Data-Management-in-Public-Procurement


 The following text is (almost) the transcript of the keynote speech I delivered at the 2018 Global Public Procurement Conference organised by the Interamerican Development Bank, Washington 17-19 September 2018



I grew up in small town in Southern Italy. In that environment the social structure can be hard to follow, even for locals. It made me curious to learn how the world works. Specifically, how we’ve made it work for ourselves-how is it that we’ve built infrastructure like roads and electrical grids? And shared community resources like hospitals and schools? The short answer is: public procurement. 

The way governments buy services and spend money for their citizens affects every one of us. Public procurement has been a part of government for hundreds of years! Engineering achievements such as the Roman Aqueduct of Segovia built almost 2,000 years ago give us a sense of what we mean by a civil work that has been properly done.

It’s easy to criticise public works. The viaduct "Morandi," named after of the Italian engineer who designed it was a major commuting link in the north-west of Italy. It collapsed at around noon on the 14th August killing 43 people. The bridge was only 51 years old.Investigators have already pointed out that alerts on structural damages had been repeatedly ignored. 

How did this happen? Simply put, poor communication between the government and the concessionaire’s safety team. Information about crucial structural aspects has been badly processed and possibly ignored. Information and its management are the keys to smart, functioning public procurement-just like those Roman aqueducts.

This is where my work comes in: In high school, I realized that math and statistics are excellent tools for modeling the uncertainty of our societies. Models trying to capture different aspects of uncertainty are complex objects, and may fail. It can be daunting. But it can also be thrilling-becauseit’s our lives! Life is the ultimate laboratory. Although we may fail to understand exactly what kind of information and management systems we need, our failures provide us opportunities to learn. Bad prediction isn’t bad news-it’s a way of making your thinking finer and more precise. 

"We are trying to prove ourselves wrong as quickly as possible, because only in that way can we find progress.
Richard P. Feynmnan (Physics Nobel Prize Winner)

And that’s why diving into the world of public procurement was the best way to carry out my childhood dream: It showed me how our societies work, and what they need to thrive.            

Close your eyes and picture a bridge, a road, a building. It’s easy to see these structures on their own. Now think about how they are all connected: In this case, the road paves the way to the bridge upon which the employee drives from her home to her office building. Both the road and bridge matter as parts of the transportation as well as the energy distribution system. They also play a role in her and the community’s safety. The road and the bridge provide services whose features change according to users’ behavior. This is a smart city.

Some of us are already living in smart cities where buildings, roads, transportation system are conceived to provide a variety of connected services.  These cities work so well because they have a wealth of information.

There are a lot of different types of information. It’s how many vehicles the road needs to serve, at what time of day, and how many people drive in those vehicles. It needs to be continuously collected and processed in order services to best respond to users’ behavior. Cities have to become smarter both in developed and developing countries. The next challenge in public procurement is managing and upgrading these smart cities. "Road maintenance" includes roads, street lighting, and electricity distribution. Information will drive the nature of procurement in an interconnected world.

Collecting this information is a challenge for two reasons. First, we need to know what data we need, and second, how to get it. Again, it’s a daunting task-frightening even! 

In 2014 the World Bank hired me to complete two missions for the Republic of Ethiopia. I was sent to the country to assess the federal centralized procurement system. A crucial meeting had been scheduled with three top officials. These men were supposed to help me find the savings the government had put away by bundling common-user items purchases in bigger contracts. I explained to them the pieces of information I needed to draw some conclusions. They listened to me attentively, then the most senior one looked at me as if I were an alien. He asked: "Why spending all this time hunting for figures? We can tell you what happened." He was afraid to give me the data because he was worried that I would draw different conclusions than they had. But that was exactly what I was sent to do! The ultimate purpose of my mission was to help them see things from a different perspective. 

Relying on too little data is one problem in public procurement. The other happens when we rely on them blindly. It’s not enough to have information alone; we need to use our common sense to trust and interpret it. 

Tyler Vigen, a law student at Harvard, set up a fascinating website with an accompanying book on many circumstances where one can find extremely strong statistical correlations among phenomena, which are extremely unlikely to be logically related. For example:


Intriguing, isn’t it? Who among us would have bet on such a positive correlation between "Per capita consumption of mozzarella cheese" and "Civil engineering doctorates awarded"? And yet, we all know that there is no actual relationship between these two variables. 

The "statistics of wishful thinking" is the danger of jiggling with data so as to impress others (... hum ... say, policy makers). Unfortunately, this happens more often than you think. During the past few years, I’ve attended over a dozen international meetings on the benefits of using e-auctions in public procurement, the so-called reverse auctions. Believe it or not, I have never heard any speaker presenting figures on the savings from reverse auctions below 50%! I’ve come to the conclusion that, perhaps, in public procurement any figure below 50% is just a curse... (or maybe forbidden!)

In another occasion, I was involved in a high-level meeting on the benefits of centralised procurement in the European Union. In other words, governments buying in bulk. Representatives from centralised agencies were presenting different models in drastically different countries, with obviously different social needs. But guess what? There was a weird coincidence in terms of basic performance: Each agency was able to achieve savings exactly between 19% and 21%! (fortunately, in centralised procurement the below-50%-threshold curse does not apply) Virtually, the same performance across different countries and systems. It was like magic!

The statistics of wishful thinking works to me as a sort of antidote against the "statistics of deception." True, sometimes figures show something we may not like. They may not disappoint as much as the kind of evidence mentioned by the character Albert Burnside in the 2008 movie "Nothing But the Truth":

"A man can live a good life, be honorable, give to charity, but in the end, the number of people who come to his funeral is generally dependent on the weather."

Data don’t lie. They are the key to using public procurement to advance our lives.     Evidence-based decision-making in public procurement is becoming increasingly urgent. 

We have examples of how good governance has led to thriving communities through public procurement. Take the project "Iscol@" promoted and funded by the Region Sardinia in Italy, which ultimately aims at drastically reducing school drop-out rates and to improve the quality of human capital. One of the pillars of the project consists in refurbishing school buildings and constructing brand new ones so as make them "modular spaces". Internal physical features such as the size of the rooms can be modified according to the subject taught. This is really smart, isn’t it?   

In a not-so-distant past we seemed to be comfortable in undertaking public procurement reforms without looking at evidence before making crucial decisions. We have seen how this can go awry, as with tragedy of the Morandi viaduct. My understanding is that policy makers were satisfied with idea that systems could be reformed by the force of principles and analogies-without actually learning from experience.  Transparency, for instance, is "holy grail" in public procurement. Analogy has been used when mimicking allegedly self-evident successes in the private sector such as the e-commerce platforms and, guess what, reverse auctions.

Times have changed already. The buzzing of procurement activities worldwide goes hand in hand with information production, which gets disseminated in multiple forms. We need information to learn from the past, to correct our mistake, and to move forward. Information is sometimes scattered-owned or managed by different organizations. Worse, information about the same phenomenon is written in different ways, which can be confusing. Our learning ambition is all too often frustrated by the babel of codes to describe the same thing.

Here is the good news: It’s becoming easier and easier to gather and correctly interpret information for public procurement. The Data & Analytics Framework experiment in Italy is one such example. At its core, it is a partnership between a big data platform and two teams. The data platform acquires, processes and provides information for analysis and machine-to-machine interchange. The teams are comprised of data scientists and visualization experts. Together, they use visual storytelling to effectively convey the information contained in the data. It will become possible to connect data from different sources that relate to different aspects of social life that are not independent from each other. 

Close your eyes again and visualize a street in a city... water pipes, fiber optic cables and metro tunnels under the surface...tramway tracks, lighting infrastructure above...a massive amount of data and information is produced about different aspects of our lives: transportation, safety, and health. The same street can be looked from different perspectives, which requires integrating different sources of information. 

Only a few years ago this would have been unimaginable. The potential of the Data & Analytics Framework as well as of other projects in same spirit is not limited to better exploit information on the procurement cycle of a vital infrastructure. It is a new way of using information for new forms of government, new forms of participation in the management of the res publica. It is the door leading to a new form of citizenship-all possible by tackling the difficult questions of information gathering and management head on.

And if you wonder whether this new scenario will be populated by as many lawyers as in today’s procurement, well the answer is... YES!

New professionals will join the team...with original thought, clothes, and maybe even tattoos.

Thank you!


1
Create a website