Under T2S, many banks will expect to benefit from needing only a single pot of collateral in order to trade across Europe, using a harmonised set of rules and processes. Many assume that connecting with T2S will reduce the amount of collateral required thanks to the creation of an effective single market, furthermore that the role of the agent bank will be reduced to little more than a connectivity provider.</p> But the numbers suggest otherwise; while direct involvement in T2S may offer advantages for certain firms, these advantages can be significantly increased through the use of an agent.</p> The original expectations for cost reduction under T2S took their first hit when the UK, the Nordics and Switzerland chose not to join the project. The expectation now is that European banks will be able to net collateral off into a smaller pot to use across the whole Eurozone rather than using separate pots for each national market, liberating assets for use elsewhere. Where some banks and brokers are trading short and others long across different markets, firms can net off the settlement to free up liquidity further.</p> This makes for good theory; however an analysis of data conducted by BNP Paribas Securities Services indicates that the liquidity optimisation provided by T2S for an individual bank, broker, or global custodian becoming a directly connected participant (DCP) to T2S or connecting indirectly via a CSD or ICSD is far from certain.</p> Let’s examine the two basic assumptions behind this theory.</p> Assumption 1: “Bringing 24 markets together, T2S will decrease the collateral required to manage liquidity”</h3> This makes sense when the effects of cash pooling are taken into account but is not otherwise certain.</p> Expectations of intraday liquidity benefits for DCPs rely mainly on the auto-collateralisation mechanism (where T2S automatically collateralises liquid securities in a nominated account to cover shortfalls against settlement obligations in other accounts) and the use of single cash accounts across all T2S markets.</p> The benefits of auto-collateralisation for intraday liquidity are limited to bonds, however, and will not solve overnight liquidity issues when financing long securities positions. Figures 1, 2 and 3, below, estimate the intraday liquidity requirements for various participants in a pre-T2S environment (liquidity managed at each market level) versus in a post-T2S environment (liquidity managed regionally as if one single market).</p> The analysis uses one month of transaction data for each type of participant (bonds and equities flows). The graphs show the relative estimated central bank liquidity usage for 3 types of clients that may decide to self-settle, in a pre- and post- T2S environment.</p> It’s clear that the benefits are not as dramatic as many may have assumed.</p> fig 1. Equity broker pre- and post- T2S estimated central bank liquidity usage</p> fig 2. Investment bank pre- and post- T2S estimated central bank liquidity usage</p> fig 3. Global custodian (equities and bonds) pre- and post- T2S estimated central bank liquidity usage</p> </p> Trading strategies are more regional than domestically oriented: the same investor is usually not taking long positions in one European market versus short positions in another European market.</p> Assumption 2: “T2S will remove the need for the agent bank model”</h3> Agent banks will still add a lot of value through asset servicing and settlement, services which are strongly linked and nuanced from market to market. Liquidity provision by agent banks could however be challenged with auto-collateralization being extended to the 24 T2S markets, but auto-collateralization will be made available for only certain types of bonds and will not cover overnight financing needs.</p> In addition, this assumption does not take into account liquidity optimisation that can be generated by an agent bank taking intraday opposing positions (long versus short), reducing liquidity requirements for market participants at the agent bank level, further, when common clients trade against each other agent banks do not need to inject liquidity. The red lines on figures 4 and 5 show our relative estimated central bank liquidity usage as an agent bank settling our clients’ trades in the main T2S markets, the green line shows the same for all market participants if directly connected to T2S.</p> The differences are significant.</p> fig 4. Liquidity requirements for settlement of bonds</p> fig 5. Liquidity requirements for settlement of equities:</p> Going further we are able to assess liquidity optimisation depending on the number of clients an agent bank has. Figure 6 shows the theoretical liquidity requirements for self-settlement versus liquidity requirements when using an agent bank.</p> This uses trade data from existing clients with various trading profiles, normalized trade data so that intraday liquidity needs are roughly the same for each client.</p> fig 6. Liquidity requirements for self-settlement versus liquidity requirements when using an agent bank</p> In conclusion, T2S will definitely help to optimise collateral requirements thanks to the auto-collateralisation mechanism but for intraday liquidity management and bonds only.</p> Agent banks will continue to create true value in asset servicing and settlement, but also in the liquidity field, taking advantage of client diversification, an effect that firms will not achieve using their own investment streams. As such participants using agent banks will see much better intraday funding compared to those that do not.</p> T2S creates a new model for settlement in Europe but the apparent benefits do not always stand up to scrutiny. Banks, custodians and brokers need to speak with their partners to test any assumptions they have made about how to reap the greatest benefit from this game changing upgrade to Europe’s financial infrastructure.</p>