Three things we learned about the Lynx payment system

Introduction

Every day, households and businesses in Canada use payment systems to send millions of transactions worth billions of dollars. In recent years, Canada has been modernizing its payment ecosystem to ensure it will continue to meet the needs of Canadians in the digital economy (Miller and Olivares 2020). This modernization includes replacing the electronic system used to transfer large payments between financial institutions—an initiative led by Payments Canada (the operator of Canada’s core payment systems) in collaboration with financial institutions and authorities.

Payments Canada achieved a key milestone on August 30, 2021, when it launched the new system, called Lynx. Lynx replaces the Large Value Transfer System (LVTS), which served as Canada’s high-value payment system for more than 20 years. Upon Lynx’s launch, the Bank of Canada designated it as a systemically important payment system, bringing it under the Bank’s oversight (Bank of Canada 2021).

Lynx uses a real-time gross settlement (RTGS) model, which is today’s global standard for wholesale payments (Bech, Shimizu and Wong 2017). In contrast, the legacy LVTS accumulated its payment activity over the business day and then settled the net amounts owing between participants at the end of the day. This caused intraday credit exposures between the participants that were managed by LVTS risk controls, and the system was ultimately backed by a guarantee of settlement from the Bank of Canada (Arjani and McVanel 2006).

In an RTGS system, there is no credit risk between participants. A typical RTGS system also settles payments in real time, but it requires participants to provide more liquidity. This is the traditional trade-off between credit risk, settlement speed and liquidity efficiency.

However, Lynx is equipped with other tools that help lower the liquidity needs of financial institutions, without reintroducing significant delays (Kosse, Lu and Xerri 2020; Bank of Canada and Payments Canada 2022).

Suppose two financial institutions each need to send the other a large payment. Both institutions can submit their payments to a queue without providing liquidity. Algorithms in Lynx look for queued payment activity that can be offset. Finding these offsets lowers the liquidity needs for participants, though it does impose a moderate delay in settling the payments.

How has the transition to Lynx changed Canada’s wholesale payment patterns? What is its effect on the efficiency of Canada’s wholesale payment system? Desai et al. (2023) provide a quantitative assessment of these questions in a recent study of 6 million settlement transactions registered during the six-month period starting in October 2021.1 These observations are compared with settlement data from the legacy LVTS during the equivalent six-month period in the previous year.

The analysis of the settlement data from LVTS and Lynx reveal three key findings. Compared with the LVTS:

• Lynx handled larger aggregate payment flows, but with no significant change in seasonal patterns.
• The bulk of the daily payment value was sent earlier in Lynx, and the settlement delays imposed by the queue were minor.
• Lynx achieved higher liquidity efficiency in payment settlement.

Let’s explore each of these findings.

Key finding 1: Flow of payments

By comparing the two sample periods, we find that Lynx settled a higher value (dollar amount) and volume (number) of payments than the LVTS. Chart 1 shows the annual increase for both value and volume in Lynx was about 10.5%. On its own, we would not expect this change in systems to have a significant impact on payment flows. Instead, this growth is likely due to external trends in payment activities.2 Ultimately, the usual seasonal patterns in weekly values and volumes remained similar across the two systems.

Chart 1: The flow of payments through Lynx was higher than in the LVTS

Chart 1: The flow of payments through Lynx was higher than in the LVTS

System-level daily average values and volumes of payments settled in LVTS and Lynx, weekly average

Key finding 2: Timing of payments

Bank staff examine the proportions of total daily payment value sent to the two systems by a certain time of day. During the two six-month periods, the bulk of the payment value arrived earlier in Lynx than in the LVTS (Chart 2). This earlier timing was likely because Lynx’s design encourages financial institutions to send payments earlier in the day to increase the chance of finding liquidity efficiencies. Payments that were not in the queue in Lynx were settled in real time, while payments in the queue faced a relatively minor average delay in settlement of 27 minutes.

Chart 2: Payments arrived earlier in Lynx than in the LVTS

Key finding 3: Liquidity efficiency

An important measure of payment system performance is the liquidity efficiency ratio (LER), which is the value of payments settled for every dollar of liquidity used. A higher LER means more liquidity efficiency in the payment system. During the two periods studied, the LER of Lynx was higher than that of the LVTS (Chart 3). This improvement in liquidity efficiency can be attributed to the liquidity saving mechanisms in Lynx and their wide adoption by participants (see Rivadeneyra and Zhang 2022; Garratt, Lu and Tian 2023). In addition, the design of Lynx allows participants to centralize all liquidity and payments in the same place, unlike the LVTS, which separated them into two tranches.

Chart 3: Lynx provided more liquidity efficiency than the LVTS

Conclusion

Lynx is a critical system for transferring high-value payments quickly and safely between financial institutions in Canada. Since its launch, Lynx appears to achieve a good trade-off between credit risk, settlement speed and liquidity efficiency.

Traditionally, a common concern with an RTGS system is that it requires more liquidity, though it does settle payments in real time and remove credit risk between participants. Results show that Lynx makes efficient use of liquidity while quickly processing payments in the queue. This sound system design and the strong collaboration between Payments Canada, financial institutions and authorities have helped make the transition from the LVTS to Lynx a success.

Endnotes

References

Arjani, N. and D. McVanel. 2006. A Primer on Canada's Large Value Transfer System. Ottawa: Bank of Canada, March 1.

Bank of Canada. 2021. “Bank of Canada Designates Lynx as a Systemically Important Payment System.” Press release, Ottawa, Ontario, September 1.

Bank of Canada and Payments Canada. 2022. An Overview of Lynx, Canada’s High-Value Payment System. Ottawa, May.

Bech, M. L., Y. Shimizu and P. Wong. 2017. “The Quest for Speed in Payments.” BIS Quarterly Review March: 57–68.

Desai, A., Z. Lu, H. Rodrigo, J. Sharples, P. Tian and N. Zhang. 2023. “From LVTS to Lynx: Quantitative Assessment of Payment System Transition.” Bank of Canada Staff Working Paper No. 2023-24.

Garratt, R., Z. Lu and P. Tian. 2023. “How Banks Create Gridlock to Save Liquidity in Canada’s Large Value Payment System.” Bank of Canada Staff Working Paper No. 2023-26.

Kosse, A., Z. Lu and G. Xerri. 2020. “An Economic Perspective on Payments Migration.” Bank of Canada Staff Working Paper No. 2020-24.

Miller, P. and A. Olivares. 2020. “A Road Map to Payment Systems.” The Economy, Plain and Simple. Bank of Canada, July 6.

Rivadeneyra, F. and N. Zhang. 2022. “Payment Coordination and Liquidity Efficiency in the New Canadian Wholesale Payments System.” Bank of Canada Staff Discussion Paper No. 2022-3.