How can the UK prevent another Carillion crisis?
With its poor data and dependence on ‘super suppliers’, the UK Government is sleepwalking into a situation where uncompetitiveness is becoming the norm in public procurement, argues Ian Makgill of Spend Network and Open Opps in a blog post adapted from a thread on his Twitter account.
After sharing my take on how bad the collapse of the construction giant Carillion really is, I promised to go through what the UK Government could do to avoid finding itself in a similar bind in the future. I’m writing as someone who has been working with procurement and public spending data for 15 years. I’ll start by showing you why we’re failing to prevent crises like Carillion and finish with some ideas for improvements.
The biggest problem that we face is that more and more money is going to fewer and fewer suppliers. In local government, 20 companies earn more than the 75,000 small businesses that supply the sector.
This creeping consolidation has been happening for many years, long before this government came to power. It is giving small clutches of “super suppliers” near monopoly status. We’ve analyzed the spend of 300 public bodies, between 2011 and 2016. Just three of them had no trade at all with Capita, another major outsourcing firm. We’re becoming dependent on firms like this.
We know it is common for suppliers to provide services for free to departments in advance of a big project. This apparently charitable offer actually gives them a huge advantage in the upcoming higher value tenders.
Competition isn’t just a UK problem. The Economist looked at tenders across Europe and found that the total number of tenders receiving just one bid rose from 17% in 2006 to 30% in 2015.
Now, I am not suggesting that there’s a structured plan to reduce competition. It’s a creeping, collective failure that means we’re sleepwalking into a situation where uncompetitiveness is becoming the norm.
This is because we’re not paying attention to the large-scale changes that are happening. The devil is in the detail here, and some of the things we’re doing are pretty worrying. For example, some very large tenders have been issued with worryingly short timescales. And some of the tenders that are being published aren’t giving suppliers enough information to even find the opportunities. As before, this creates a failure of competition.
If you’re on the ground, letting contracts for a council or a university, you can’t see a creeping change in competitiveness. But if you take a macro view of the market and put together the data from multiple sources it is startlingly clear that we have a problem.
There’s something genuinely wrong here, but solving this isn’t about politics or setting new policies. For instance, we have very strong policies for awarding contracts to small and medium enterprises (SMEs). They don’t seem to be working: 23% fewer contracts were awarded to SMEs in 2017 than in 2016, according to data on the Government’s Contracts Finder site.
It’s not necessarily the buyers that are failing. They have a very difficult job with competing priorities. They’re doing the job that they’re required to do. At Open Opps, we can take that wider macro view and it shows we’re giving a significant advantage to incumbent suppliers or suppliers with a long history of trading with the public sector.
So how to fix this? What we have to do is relatively simple, but making it stick will be hard.
Firstly, we need much, much better data. If you have good data, decisions are easy. We’re currently working with our hands tied behind our backs. If you want to know who won a contract, the chances are that you can’t find out. If you go to Contracts Finder, the central repository for public tenders, you’ll find 28 contracts mentioning Carillion. That’s just 5% of the estimated total number of Carillion contracts.
How can we fix this problem if we don’t know where the contracts are? In 15 years of working in public contracting, I can count on two hands the number of organizations that are confident that they have an up-to-date register of all of their contracts. In most UK public organizations, there is no real understanding of the contracts that they have in place, their value and when they end. I previously worked on an excellent counter-fraud project set up to analyze fraud in public procurement. That project achieved less than it should have because buyers felt they weren’t allowed to share data with each other
We need an honest conversation about how to improve the data we have on contracts. Each sector within Government thinks that they just need to understand their own spending, but each of them is trading with this clutch of super suppliers. We can’t afford to monitor contracts on a piecemeal basis, we need a view of the whole market.
Secondly, this data needs to be open. This is not a nice-to-have, it is a critical part of repairing the imbalance between incumbents and the businesses and it will drive forwards competition.
The public also needs to know much more about the performance of these contracts. We should be able to see every contract awarded to a supplier across all parts of Government in an instance.
At some point in the last few weeks, as Carillion began to fail, someone in the Treasury would have asked: “How bad will it be for the government?” The only honest reply would have been “we don’t know”, and this is simply because staff in public bodies are simply not bothering to upload information on which company has won which contract.
We should be able analyze past performance and to monitor overspend daily. This holds buyers and suppliers to account and shows whether suppliers are performing properly. We should be able to see when buyers are routinely breaking budgets and when suppliers are routinely expanding contracts beyond the budgeted value.
Transparency will allow us to independently verify the progress of these contracts and the whole market. If the analysis is left solely to the Government, we’ll just get a bunch of Soviet-style tractor production statistics.
Finally, we have to set up a program of improvements based on the data. We need to watch the spread of the largest suppliers and to monitor their risk. But also we need to see when buyers are letting uncompetitive tenders.
With good data, we can know what works. For example, we can test what factors are engaging SMEs and which factors are preventing them from bidding on tenders.
Effectively, we need a much more scientific, data-led approach to find out how to deliver the key priorities for our public contracts. It doesn’t matter if they’re large or small, we should be watching and nudging them to improve.
I was in Norway recently attending the launch of an EU-funded project called “They Buy For You,” which has been set up to do just this. It will test and analyze public procurement data across Europe to help governments improve their purchasing. We were just 20 people in a room, but we were confident that we’d be able to gain new insights for a whole continent’s contracting. So this doesn’t have to be hard.
Open Opps has already built many of the tools we need to succeed in this space (often based around the Open Contracting Data Standard). Government will say “it’s too hard” or “we can’t afford to monitor all of our contracts” but the reality is that we can’t afford not to.
Photo: Elliot Brown: Abandoned Carillion crane (CC BY-SA 2.0)