PDD#02: Payment System Structures

PDD#02: Payment System Structures

In the first post we looked at how the payment processing, as we know it today, came into being. We saw the transition from barter trades between two parties, to an intermediary based payment transaction of modern world. Now before we jump into the structure and economics of any one specific payment rail (like Card payments, ACH, RTP, FedNow etc.), let us look at how the entities are structured within the anatomy we defined in that first article. In this post, I aim to breakdown how that intermediary based structure expanded to the multiparty transactions structure that we have today.

This post will focus on:

  • Various payment system models, based on how they are organized
  • Payment value chain, and it's disruption due to internet


Payment systems, based on how they are organized

Almost all countries have their commercial banking system built under the central bank. The central bank is hence called bank of banks, and is also a supervising means of country's payments and inter-banking clearing systems.

When we pay using anything other than cash, what we are actually spending is the bank deposit. The bank deposit is the money deposited in the bank, and is managed via accounts and ledgers. The central ledgers are held at the central bank level.

Say you paid for your lunch via peer to peer app (for example Zelle in US, Paytm in India). What you have paid the merchant is a bank deposit from your bank to their bank. When they both eventually settle, central bank will need to move cash from your bank to that of merchant's bank. Now of course they do not do this for every transaction, but that is the simplistic view of how the settlement is happening post your lunch transaction.

No alt text provided for this image
When we spend the money, we are spending Bank deposits. The final settlement happens in cash eventually


The centralized system is also called as the trust based system. And you can now add layers and layers of intermediaries to facilitate the access of funds and the various payment use cases. Within this centralized setup, a payment can be processed within a closed loop or an open loop:

1. Open Loop - Centralized Systems: Open centralized payment systems are operated by a central authority or entity, and they allow participation from multiple users and organizations. They usually follow the hub and spoke model. These systems are open to various participants, including banks, merchants, and consumers. Examples of open centralized systems include traditional card networks like Visa and Mastercard, where a central entity manages the payment infrastructure and facilitates transactions between different parties. The check and electronic funds transfer payments are also open loop.

To run an open loop system, the centralized entity would have to detail how the parties would interact, what would the successful flow look like in terms of experience and messaging, what does failed flow look like, and in the event there is a dispute what process would be followed. The eventual arbitration would be decided by the central entity.

No alt text provided for this image
An Open Loop Payment System (centralized)

2. Closed Loop - Centralized Systems: Closed centralized payment systems are also operated by a central authority or entity, but in this they have direct relationship with one or both end parties. PayPal, Western Union are all closed loop systems.

American Express cards mostly work within a Closed loop too. American Express has relationship with both customer (they issue their own cards, unlike Mastercard or Visa) and they also work with merchants directly (again, unlike Visa and Mastercard). This has shifted a bit in last decade though, as they spread out to work with acquiring companies, but still the higher value merchants have direct American Express relationships. We will discuss more on this when we look at card payments.

No alt text provided for this image
A Closed Loop Payment System (centralized)

Another example is a closed-loop payment system used by Disney at their theme parks, where visitors can make purchases within the park using a Disney Magic Bands that are not accepted elsewhere.

No alt text provided for this image


To process payments outside centralized systems, one need to first solve the issue of "trust". In 2008, a whitepaper titled "Bitcoin: A Peer-to-Peer Electronic Cash System" authored by?Satoshi Nakamoto did exactly that. The paper introduced peer-to-peer version of electronic cash that would allow online payments to be sent directly from one party to another, without going through any financial institution. That concept was based on an improved and truly decentralized blockchain. That opened the innovation floodgates for the decentralized systems, to say the least.

No alt text provided for this image
Decentralized payment system

3. Open Loop - Decentralized Systems: Open decentralized payment systems operate on a decentralized network without any central authority controlling the system. These systems allow participation from multiple users and organizations, and the processing of transactions is distributed across the network. Cryptocurrencies like Bitcoin and Ethereum are examples of open decentralized systems, where transactions are verified and recorded by a distributed network of computers (nodes) without the need for a central entity.

4. Closed Loop - Decentralized Systems: These are similar to closed loop centralized systems, the only difference is that they operate on a decentralized infrastructure. An example could be a private blockchain-based payment system implemented within a consortium of banks, where transactions are recorded and validated by the participating banks without involving external entities.

No alt text provided for this image
MANA (Decentraland currency) is a closed loop payment type



Payment Value Chain

A payment value chain refers to the sequence of activities and entities involved in facilitating a payment transaction from initiation to settlement (and clearing). It represents the flow of value, information, and funds between various participants in the payment ecosystem.

The Payments value chain, similar to value chain for any given consumer market, can be seen as comprising of three parts:

  1. The consumers, merchants or users, who are asking for the payment service
  2. The suppliers or financial institutions that provide various financial products or services
  3. Distributors - The intermediary glue that acts as a middleman. They work with the suppliers or financial institutions in our case, and can offer the benefit of pooling risk, reducing cost, and providing economies of scale, among others.


The financial world follows certain rules of economics, which you can see reflected in payment value chain too:

  • Ad valorem pricing, instead of flat (unless flat pricing is more economical)
  • The receivers usually pay
  • More open a system, larger the scale it needs
  • Any exception processing costs extra
  • Risk finances the flow, and the entity who is taking largest risk, gets the largest share of pie

I will dive deeper into this when we talk about specific payment method.


Internet's impact on Payments Value Chain

When internet came and allowed the digital information to be exchanged at high speed and almost zero cost, it ruptured value chains across the industries. Payment systems were no different and saw it's own disruptions.

If there is a single phrase that describes the effect of the Internet, it is the elimination of friction. With the loss of friction, there is necessarily the loss of everything built on friction...


Above quote is from Ben Thompson (famous author of strategy and business analysis). In 2015 he introduced Aggregation theory to explain how this loss of friction impacted the value chains:

The best way to make outsize profits in any of these markets (suppliers, distributors, and consumers/users) is to either gain a horizontal monopoly in one of the three parts, or to integrate two of the parts such that you have a competitive advantage in delivering a vertical solution. In the pre-Internet era the latter depended on controlling distribution.

Although he noted it in context of consumer market, that model is very relevant in the generic sense too. In financial world, the suppliers are providing financial services like consumer payments, consumer credit (lending), bill payments, wealth management (stocks, bonds, alternate investments), payroll processing etc.

Previously, the banks were sole centralized players who bundled together payrolls, bill payments, home mortgage, wealth management and other products. With advent of internet, many new niche players with much better user experience came up and started unbundling the financial space. These "FinTech" players focused on solving the customer pain points, that the banks earlier did not have to bother much about. Take an example of how Bank of America (BofA) today offers APY of 0.01% even as Fintech platforms like Marcus are flashing 4.15% APR on savings account.

No alt text provided for this image
BofA rate of 0.01% APR vs Fintech rates of 4.15%+ (July 2023)


They do this by focusing on their niche from the unbundled use cases of a traditional bank. As Panagiotis Kriaris noted in his post from 2022:

"...where under the old model, banks used to be the sole provider for both the customer-facing and the back-end set-up in a bundled mechanism, the evolution of Fintech has allowed the (partial) decoupling of FS offerings from outdated rails and?banking?infrastructure. This has,?in turn, enabled taking control of the user interface and has unleashed a huge wave of new applications that were not bound or restrained by old legacy systems."
No alt text provided for this image
How FinTechs have launched products in various areas of BofA's bundled services. Source:Panagiotis Kriaris article, graphic based on CB Insights article


Then there are other companies like Intuit, who have focused on building a vertical solution by aggregating all these banking and fintech accounts of customers into a single consolidated view. They then integrated their tax preparation services, credit score tracking service and even ad platform on top of this consolidated view.

You can see how Intuit's move from offline QuickBooks software to vertical solution has allowed them to grow massively and also effectively tap into the power of AI. Even at the scale of USD14 Billion annual revenue, the company is able to command a 78% gross margin (22% operating margin)

No alt text provided for this image
Intuit holding aggregation
No alt text provided for this image

Closing Remarks

We ended last post with breakdown of the payment anatomy.

In this post we have now established a model for classifying payment methods - based on whether they are centralized or decentralized, and whether they are open loop or closed loop. We have also broken down the entities in the payment value chain, and looked at the Aggregation Theory of Ben Thompson to help us understand how technology contributed to reshuffling of value chain, due to elimination of friction.

Next we will pick up one payment method and use these tools from two articles to understand it holistically.

Anand Sundaresan

Principal Architect | Presales | IT Strategy | Enterprise Architect | Payments Solution Architect | SAP Architect | Data Architecture | Offering 30+ years of IT experience in various industries | SC Cleared upto 09/2029

1 年

Good article Jasginder Singh, simple and ??. This will also help with understanding for everyone including those who are wondering how their QR scan payment from a mobile at the vegetable vendor shop happens through the intricacies of banks clearing and settlement systems. Also to wonder on how banks can just fall off when liquidity crisis happen. Hope to see in a later article keep it up

Nishtha Arora

Director, Product | Scaling B2C/B2B SaaS (0-> $50M Revenue) | Gen-AI & ML Innovation | Ex-Walmart, Google, Airbnb-backed | 13+ Years Driving Scalable Solutions

1 年

really like the analogy you've presented for centralized and decentralized banking. It made me think about the merits of each approach in a different context. Bear with me :P In centralized banking, it's like having a teacher oversee students in pairs during quizzes. If one student gives the right answer, the other student gives them chocolate. This ensures that the teacher is the sole judge of the answers, but there might be concerns about the teacher taking too much chocolate for themselves in between to reward herself of fair evaluations (sorry teachers :D) On the other hand, in a decentralized banking approach, it's like students monitoring each other without a teacher present and a pair is being monitored by the whole class. In this case, clear communication between all students is crucial, and maintaining privacy regarding who gave right or wrong answers is important (students need to wear masks :) ). Also, all students want to be top rankers. What if they all decide to give marks to the best student :D. We want to avoid a situation where teachers have access to too much of our chocolates but privacy & data maintenance & networks becomes a huge questions in decentralised banking.

Prasanth Raja

Managing Partner-Trusted Advisor of Digital & Business Transformations | Visionary & Multi-Skilled Leader

1 年

Good article Jas! Please continue sharing the wealth of information you have in the payment systems.

Abhishek Prasad

Analytics Consultant

1 年

Thanks for posting. Explained very nicely Jas .:)

Trayambkeshwar Vatsa

Chief Technology Officer @ NCDEX e Markets Ltd. | Leading IT Delivery, Project Management, Cyber Security, Enterprise Architecture and IT Infrastructure

1 年

Simple , Informative and Crisp article. Well done

要查看或添加评论,请登录

Jasginder Singh的更多文章

  • PDD#21: Tokenization of Real World Assets (RWA)

    PDD#21: Tokenization of Real World Assets (RWA)

    Last November I wrote about Payment Tokenization in PDD Article #9, which explained how Payment tokenization transforms…

    3 条评论
  • PDD#20: Buy Now Pay Later explained

    PDD#20: Buy Now Pay Later explained

    Preface As digital commerce evolved, so too did the needs of consumers, creating fertile conditions for FinTech…

    1 条评论
  • PDD#19: Open Banking - A Conceptual View

    PDD#19: Open Banking - A Conceptual View

    Note: I have been finding it hard this year to meet my biweekly commitment to write the next edition of newsletter…

    1 条评论
  • PDD#18: Payment Orchestration

    PDD#18: Payment Orchestration

    Preface In the 2010s, the digital payments were largely dominated by a handful of methods, such as credit cards…

    6 条评论
  • PDD#17: FICO Deep Dive

    PDD#17: FICO Deep Dive

    Preface The last article in this series focused on payment risk management, and how decision making models like that of…

  • PDD#16: Payment Risk Management, FICO's growth

    PDD#16: Payment Risk Management, FICO's growth

    Preface In the past articles of this newsletter, my focus has primarily been on analyzing payment structures of modern…

    5 条评论
  • PDD#15: How to build a Neobank from scratch

    PDD#15: How to build a Neobank from scratch

    Preface The idea of this newsletter came after realizing that while there was a lot of content available on the…

  • PDD#14: How RTP (TCH) rail works

    PDD#14: How RTP (TCH) rail works

    Preface In today's world where the cost to send an incremental piece of information is near zero, the immediacy and…

    12 条评论
  • PDD#13: FedNow explained

    PDD#13: FedNow explained

    Preface The payment landscape is evolving at a rapid pace, with more and more options being presented to customers…

  • PDD#12: Building Payment Apps in Hours (Using ChatGPT, Figma, Stripe)

    PDD#12: Building Payment Apps in Hours (Using ChatGPT, Figma, Stripe)

    Preface Welcome to the 12th edition of this newsletter, the last one of 2023. And a warm welcome to the 30 new…

    2 条评论

社区洞察

其他会员也浏览了