NESTA commissioned a piece of research to SpendNetwork about startup success rate in public procurement. I have not seen the underlying data or writeup, but trust Ian Makgill to have done a good job.* The headline figures are starkly: only 3% of spend goes to companies within the first 5 years of incorporation and only 0.5% in the first two years. No wonder startups stay away from public procurement.
Looking at the graphs, here's what I find interesting:
1. Younger but larger companies are more successful than smaller ones (although it is unclear if this is a correlation or a causation).
2. The success sweet spot for smaller companies is around a decade after incorporation; for larger companies there is a huge spike in years 24-26. Is being around for a quarter century a good success benchmark?
3. Smaller and younger companies are more successful in local government procurement. I am not sure this is good, bad, or just a consequence of demand aggregation by Central Government (yes framework agreements, I am looking at you)
4. In Central Government, the second most successful sector (in terms of spend) for small and young companies are the legal services. I did not expect this and cannot think of good hypothesis why.
5. At the same age larger companies are always attracting more expenditure than smaller ones (even though larger companies make up a tiny, tiny minority of the total number of companies). Only after 3 years can smaller companies pull in the same amount of spend as larger companies do in year 1 after incorporation.
6. Expenditure with larger, older companies is on the rise and with younger ones in decline. This is a very important finding as it contradicts the claims by Central Government that more money than ever is being spent in smaller suppliers. That may be true if sub-contractors and supply chains are taken into account, but it does not appear to be the case in direct procurement spend.
7. Building up on 6, the amount being spent in younger, smaller companies is getting lower as well. Is this an effect of procurement becoming more competitive over the last few years? Or a lingering effect of the dreaded PQQ love affair?
What the research does not yet show us is the following:
1. What are the contract value(s) young companies are able to be successful? I suspect there is a power law distribution with most of the success amounting to those 3% coming from smaller contracts
2. What is the participation rate of young companies in public procurement and in comparison with the overall market?
3. What is the success rate for young companies overall?
4. Under what procedure are younger companies more and less successful?
5. Is this happening in other countries? I know that in Portugal SMEs have been more successful in recent years, but I am not aware of a breakdown by year of incorporation type of analysis.
6. Who are the unicorns? The companies being successful in year 1 after incorporation that is. And what companies make it to large in year 1? Are they subsidiaries to older, larger companies?
As for potential reasons why younger companies are not more successful or engaged in public procurement I have a few ideas:
1. Procedural complexity: Not necessarily legal complexity, as I remain bullish that it is possible to make procedures a lot simpler than they are - and I have eaten my own dog food on this.
2. Opportunity cost: Younger companies (particularly smaller ones) have limited resources and allocating them to a potential contract where their success prospects is low seems like a bad deployment of those resources.
3. Timescales involved: procurement procedures are slow. Damn slow. By definition a younger company will have more difficulties managing cashflow, so even if there is the prospect of being successful in a procedure that can take close to a year, the company will probably need to have their cashflow well managed. I do not agree with Tom Symons view that payment terms in public procurement are particularly onerous. 30 day payment is becoming more and more standardised and if anything, payment terms by larger companies tend to be worse. Or else the Westminster Government would not have included in the Public Contracts Regulations an obligation to pay undisputed invoices in 30 days to sub-contractors. Now, public sector is indeed a slow payer in other countries...
4. Experience matters: Again this is a chicken and egg type of problem - without experience is hard to win contracts (or pass the selection stage) and is damn easy to make unforced errors during the procurement procedure (something similar happens in grant applications). Bidders need to fail (and get feedback) so that they can become better in the future. By definition this takes time. It may be that if companies are disheartened with the public sector if they are unsuccessful in their first attempt and either never come back, or only do so many years later.
5. Risk aversion from public procurers: The old adage that no one was ever fired for buying IBM is still valid in public procurement. Procurement officers have a bigger downside if something goes wrong in a contract (particularly if they pushed for a smaller supplier) and barely any upside if things go ok. Their career progression can be derailed if a contract goes pear shaped, but no one will care if SMEs or younger companies were particularly successful winning contracts under his watch.
6. Analysing candidate's information at the start disadvantages younger suppliers: This is something of an hypothesis I have been mulling for some time and is based on how we form impressions and evaluations. Selection stages where candidates "qualities" are being compared and contrasted naturally leaves younger companies at a disadvantage: their financials are going to be weaker and their experience shorter/less relevant. Even if they make it past the selection stage, they may do so as the "ugly ducklings" of the procedure and if the same public procurers mark the tenders, heuristics will kick in, making it more likely that the "best tenderers" also have the "best bid". After all, there is a reason why in higher education marking tends to be anonymous... I am yet to prove this theory, but give me time, money and data.
* I pestered Ian a few months ago (or maybe last year) that this is was an area that deserved attention. I am happy that he found a way to do it.