“We need a DLT highway for the capital market”

Because we cannot afford to miss developments in the capital market. Unlike most other banks, KfW refinances itself largely through the international capital markets. We need to know what is moving there, which technologies are prevailing and whether they can help us make our processes more efficient. If we look to Asia, the US, UK or Switzerland: things are happening there. And we, too, must start testing new technologies.

One aspect that drives me is demographic change: the aging population is leading to a decline in the working population and an increasing shortage of skilled workers. Innovations and new technologies can help to compensate for this. And as an employer, we also need to remain attractive to the younger generation. We rely on innovative approaches and pioneering work for this.

We want to understand how we can use existing and new technologies to make our issuance process more efficient, without sacrificing other characteristics such as the security and liquidity of our bonds. Because these are characteristics that our investors expect.

As early as 2017, we conducted an initial test transaction using what is known as distributed ledger technology (DLT). At that time, the legal framework was still lacking, but in the meantime, the legislature in Germany has taken up the issue. In 2021 the Electronic Securities Act (eWpG) came into force and now things are getting exciting.

The eWpG created the possibility of dematerializing the physical security for the first time. In plain language, this means entering the security into a digital central or decentralized register. For the latter, distributed ledger technology (DLT) can be used, which also includes blockchain. We have tested both: the digital central register in 2022 and the decentralized register in 2024.

For the first step, we focused on using existing technologies and in 2022, together with Deutsche Börse and Clearstream, we issued our first digital bond in the form of a central register security on its D7 platform. This involved bonds worth 20 million euros, i.e. a manageable pilot project. The initial aim was to store the physical security in a digital central register instead of physically locking it up in a safe as before. In Clearstream, we have an experienced partner at our side, with whom we have learned together and optimized processes.

In a next step, we then also issued our first large volume euro benchmark bond via D7. Since then, issuance via the D7 platform has become standard for our euro benchmark bonds. By the end of 2024, we had already reached a volume of 8 billion euros on D7. And by the end of February we were already at over 10 billion euros. This shows that the issuance of central register securities has already become scalable for us, i.e. it can be increased in size at will. All our digitization initiatives aim for scalability.

Currently: no. But that was not the primary goal of our first pilot transactions. Our goal was above all to test the new possibilities that the legislation has opened up for the issuance of securities – in particular the use of DLT. In doing so, we wanted to involve as many other market participants as possible in order to learn together, identify new risks and test new roles in the overall process.

To this end, we brought a four-member banking syndicate on board. Deutsche Bank was a key partner here, it planned a long phase of market preparation, what we call market education, and approaching investors before the first transaction. And this paid off: in addition to our anchor investor German fund manager Union Investment, a good handful of other investors ultimately participated in the transaction. This has never happened before.

Some see the great potential and find it attractive to be first movers. Others say, “It's not mature yet, we'd rather wait and see.” Both are understandable. But experience shows that once the big market players move, the others follow. And we have seen that interest has increased significantly in the last two years.

Yes, quite a few! Because unlike central register securities, crypto securities rely on a completely new technology, DLT, or in our case specifically blockchain. We needed to do some educating in advance. First of all, everyone internally had to understand what it was all about. A security is a security is a security, whether on paper or on a blockchain – but the way it is processed changes radically.

Our legal department, compliance, IT – everyone had to get to grips with the new technology. You can't just transfer today's analogue processes 1:1 to the new technology. We also had to define completely new job profiles that didn't even exist before.

Of course, with every new technology come new risks, so you have to start learning early on. And then, of course, there are regulatory hurdles: who is liable for what? How does the regulator assess these new structures? These are questions that you don't just clarify in a coffee break.

A big issue is interoperability. Currently, market participants use many different platforms, but these are not all connected to each other. But as long as I don't have a connection from platform to platform, it won't work: without the appropriate infrastructure, you can't create liquidity.

Our first crypto security bond was issued via the Polygon blockchain. But that was just one of many possibilities. But as long as everyone has their own blockchain, scalability is lacking. We need a kind of “DLT highway” that everyone can connect to and move on. You can picture it: security checks such as KYC (Know Your Client) are carried out at each access point – but once the process is on the “highway”, it rushes through. We are also currently discussing this with the ECB and the major central securities depositories such as Clearstream.

Exactly, this technology is only attractive if both the security and the payment transaction can interact on the same level.

That's why, in a second round, we tested how payment transactions can also be processed on the blockchain as part of the ECB Exploratory Work. For this purpose, we used the Bundesbank's trigger solution. This was important because only when securities and payments are on the same technical level can real efficiency gains be achieved.

It worked: delivery versus payment settlement (DvP), a securities settlement procedure that states that payment is made either before or simultaneously with the delivery of the securities, was successfully executed without a central securities depository – in less than three minutes! But it also showed that there are still open questions – especially regarding necessary rules. In theory, such neuralgic points cannot be identified at all; this can only be found out through practical testing.

We are all learning every day, including me as Treasurer. What's important to me is to ask at the beginning, what should the goal be? It's not so bad if we don't achieve the goal, because then we can re-orient ourselves. In any case, you're smarter afterwards and you know: we don't have to go down this road again.

In the second half of 2025, we want to issue another blockchain-based digital bond. Of course, all our new findings will be incorporated into this. We are also continuing the dialogue with regulators, for example on the capital adequacy of digital securities. Europe is now very much at the forefront here because there is a very good regulatory framework. The topic of digital central bank money (wCBDC) as a link between blockchain systems is also becoming increasingly relevant.

The big question now is: how fast can we move forward? Since European competitiveness is always linked to this, there is a great deal of interest in us making progress. The goal is to harmonize the fragmented European market and thus create growth.

I think we are currently experiencing a natural evolution. Our world is becoming increasingly digital, because digitization ensures greater efficiency – this has also been shown in other areas. And once we have established the “DLT highway” on which everyone can travel quickly, we are talking about completely different dimensions.

There will certainly be a transitional phase in which both ecosystems – the digital and the analogue – will run in parallel. But in the long term? Think of artificial intelligence in conjunction with the potential of blockchain – there's an awful lot of potential in that. This is creating enormous dynamism in the capital market.

If we wanted to be really visionary, the next step would be a tokenized economy – you can implement things much more efficiently and quickly there. If the technology is mature and secure, why should you still go the old way?

I am not a digital native but I have a smart home, for example. If I see an advantage, if it makes my life easier – why not?

The interview was conducted in February 2025. Photos: Benjamin Krug.