John Pellew, principal, distribution and securitisation at Arrow Global, argues that it is possible to de-risk NPLs and democratise fixed income through digital securitisation

In isolation and in the absence of mitigating factors, non-performing loan and distressed assets are inherently high-risk investments, but that’s not the whole story.

While it’s true that we are dealing with defaulted loans and distressed assets, these assets are not sold in isolation or single units, and they are not sold because they have no value. They are sold because the banks are forced to sell under the Basel-type regulatory obligations. We argue there is value there; it’s just a question of how much value and how we can extract it.

Where there’s muck, there’s money
The NPL and distressed industry follows the old adage, “Where there’s muck, there’s money” - referring to all of the traditionally ‘dirty’ industries around the world, like domestic waste management and the burgeoning global recycling industry. I chuckle to myself sometimes when I describe Arrow as the waste recycler of the European financial services sector, but it’s true. We do the same job as any scrap metal merchant or plastic recycler: we buy waste, we sort it, we process it, and we transform it into something of value again and sell it back into the markets.

So, how does it work?
NPL and other distressed assets cannot be viewed in the same way as performing assets, where it’s easy to look at the repayment schedules of the loans and project out cashflows which inevitably already account for a level of delinquencies. NPLs, on the other hand, are themselves the delinquencies, so you need to look at value in a different way.

• It is a numbers game; in other words, large granular portfolios are much easier to predict than smaller lumpy portfolios.
• NPLs and distressed assets are traded at a deep discount to the face value, so buyers like Arrow are already de-risking their exposure on day one.
• Time is key. You can remediate NPL and distressed assets if you have enough time. You can collect out what you need to make the portfolio work, but the real risk is how long it takes.
• Collateral is king. Sometimes, but not always, there is some collateral of value that can be converted to cash to recoup some of the investment.

Once you break in a portfolio, you get a small but relatively consistent pool of reperforming loans, which become the foundation of a predictable cashflow. At Arrow, some of our ‘mature’ portfolios collect out at +/- 1% /pa. +/-1% accuracy.

In any other industry, this would be viewed as “very predictable” and therefore low risk. However, in the NPL space, it still seems to carry a negative perception.

Luckily for us, the maths involved does not lie and we find a dislocation between the real risk and the returns in this sector. This is good for Arrow and its investors, but a missed opportunity for the broader investment community.

So, how do we change the paradigm and change the world?
At Arrow, we’ve been securitising NPL portfolios for over 10 years, but historically we’ve suffered from the negative market and rating sentiment towards NPLs. This has forced deep haircuts to be applied to the ratings and heavy retention requirements, to the point where it’s just not economical or capital efficient for us to securitise in the traditional way. We’ve been forced to look at the problem in a different way, in an effort to improve our capital efficiency and optimise returns to investors.

So, where do we start? Apart from educating the market in order to reduce the fear of non-performing and distressed asset classes, I think we need to start answering questions like:

• Does the application of new tech like blockchain and AI change what the products look like?
• Do new world securitisations even need to look like a securitisation and, if not, then what?
• Could we slice and dice cashflows in different ways and achieve better outcomes for both buyers and sellers?
• What is the role of data in this future state?
• Can we have trust and transparency by design, versus having to retrospectively prove what we have or have not done?
• What is the role of analytics in the future product and marketplace?
• Is there an opportunity to truly democratise fixed income by offering pre-baked analytics and tooling?
• There is also the question of data asymmetry; in other words, who should get to see what and when?

So, what does the future of NPL securitisation look like?
First, let’s take a look at what we know so far. The regulatory landscape around blockchain, digital assets and AI is evolving quickly, giving rise to huge opportunities to create new products for new and existing markets.

It is now possible in many jurisdictions to build blockchain digital structures that behave and execute like ABS but are not an ABS and are therefore not regulated as such. This allows much greater flexibility to tailor solutions for specific buyers, whether it be to target the AML requirements of an insurance company or just the delivery of yield to an accredited investor.

We can now build much greater utility into the product itself. Logic would suggest that providing greater utility to investors should translate to better premiums for the issuers.

Where are the gaps?
Data is the number one challenge in the whole story - because, without data, you have nothing. A lot of the work we are doing at Arrow is around building better data services, whether that be at the collection point or in the storage systems or the tools used to analyse the data.

All these efforts are aimed at providing more accurate forecasts of cashflows. We are now focusing heavily on building more flexible and agile platforms using serverless technologies and building data services using graph data architectures instead of rigid SQL formats. The intention is to leverage machine learning and artificial intelligence to optimise our collections strategies and downstream product construction.

The last piece of the puzzle is smart people. You just can’t innovate without smart people. So, we are always on the lookout for talent in these emerging technologies across blockchain and the data sciences.

What happens if the industry doesn’t change?
I hope I’m wrong and the industry adapts and evolves. However, if the industry doesn’t innovate in the collection and use of data, and we can’t develop alternatives to traditional securitisation models that are capital and cost effective, then securitisation in the non-performing and distressed asset classes will become much slower. It will become more expensive and harder to justify, essentially preventing the ‘clean-up’ companies from doing their job.

The implication I would foresee is a significant increase in structural pressure on the whole financial ecosystem, forcing banks to tighten lending criteria to minimise default rates, thereby reducing their capacity to lend and keep the economies of the world moving forward. This would be a serious situation for the whole world economy, and it would require all of us to work to resolve it.

For more on trends and activity in the European NPL securitisation market, join SCI at our NPL Securitisation Seminar on 27 June, either virtually or in-person.