MiFID II rules clearly indicate that an asset manager can pay for research out of their own P&Ls or they can fund a research payment account (RPA) and pay for research using a direct charge to clients (the accounting method) or a research charge collected alongside execution (the transactional model). This week, we will discuss the mechanics of using the accounting method to fund an RPA.
The Accounting Method
The Accounting Method is the approach originally proposed by regulators in early drafts of MIFID II guidance. This approach was pioneered by a number of Scandinavian asset managers and is often referred to as the “the Swedish Model”.
Funding an RPA – the Accounting Method
1 Currently the firms planning to be RPA Administrators include Bloomberg Tradebook, IHS Markit, Instinet, ITG, Liquidnet, & Westminster Research Associates. 2 Research Procurement Solution Providers include Castine, Commcise, Eze, FrostRB, and Investars. All RPA Administrators are also providing clients with much of the functionality provided by Research Procurement Solution Providers.
The Accounting Method requires that an asset manager establish a separate annual research charge that would be charged directly to the client which would be used specifically to pay for all research the manager used on the client’s behalf. Using this method to fund an RPA, an asset manager would need to obtain agreement from each of their clients to pay this defined research charge as this charge would be withdrawn directly from client funds. The asset manager would then trade with every executing broker at an unbundled execution-only rate.
An asset manager funding their RPA using the accounting method would need to calculate a regular proportionate accrual of the research charge it established and communicated to its clients. The accrued charge would be physically transferred from each fund on a monthly or quarterly basis and sent to the asset manager’s RPA account.
The asset manager would leverage either internal software systems, external research procurement providers’ platforms, or RPA administrators to help them meet the research budgeting, valuation, payment, and reporting requirements mandated by MiFID II. Because the manager would be funding their research expenditures using direct client charges, they would not need to undertake the trade level reconciliation and accounting required under the Transactional Method.
Depending on how the asset manager decided to manage the research payment process, they would either make all payments themselves, or they would hire one or more RPA administrators who would manage all payments for research on their behalf.
Every broker or independent research provider used by the asset manager would receive a “hard dollar” check drawn on that RPA directly from the asset manager or RPA Administrator for the research they provided the asset manager.
Although the “accounting method” for funding research payments was the original approach presented by regulators in developing the MiFID II research unbundling rules, it has become less popular over time. In fact, today only 10% of European asset managers say they plan to levy a separate research charge to fund their research payments.
We suspect the reason for this is based on a number of factors. The first is that many asset managers have expressed a preference for using the transaction method where a research charge can be collected alongside transactions. This approach is similar to managers’ current use of CSAs to pay for research and thus is easy to understand.
In addition, many asset managers believe that asking their clients to pay a new “research fee” on top of existing management fees may be difficult to explain, particularly as many managers won’t be adopting this approach. Consequently, asset managers fear that charging a separate “research charge” may push clients to choose active managers that aren’t charging these fees or move to passive investment products altogether.
Another problem with utilizing the Accounting Method is that a number of large U.S. investment banks have resisted supplying U.S. equity research to asset managers utilizing the ‘Swedish Model’, insisting that hard dollar payments for research will force them to register as investment advisers and, among other things, make it difficult to trade on a principal basis.
Despite these challenges, we believe some asset managers may conclude that the Accounting Method is a more transparent approach to funding research payments than the Transaction Method, and it is less costly than paying for research out of their own P&Ls. Consequently, we suspect that some managers (albeit a minority) will decide to fund their RPAs by asking their clients to pay a separate research charge.