Volcker Underwriting: It’s Simple … No Need to Overanalyze

As banks face the July 21, 2015 deadline for proving their trading desk exemptions from the Volcker Rule, they have been focused on estimating the reasonably expected near term demand of customers (“RENTD”) under the market making exemption. [1] However, trading desks intending to take the underwriting exemption (“underwriting desks”) must also estimate RENTD, which is defined differently for underwriting and in our view poses fewer implementation challenges.

Underwriting RENTD is the anticipated market demand for an underwritten deal’s securities (estimated as a monetary range). The Volcker Rule requires underwriting desks to take RENTD into account when setting three limits intended to prevent proprietary trading: the size of a desk’s underwriting positions, the length of time residual positions can be held, [2] and the risks arising from certain underwriting support activities. These three limits will have to be justified to regulators and be regularly monitored by the bank. [3]

If implemented strategically, estimating RENTD and setting these limits should be fairly straightforward and doable. Given the reputational risk from failed underwriting deals and existing regulations that govern securities offerings, most banks already have (and should use) policies and procedures that are already in place. These processes effectively estimate RENTD and set limits for underwriting deals that are similar to the three limits required by the Volcker Rule.

Furthermore, existing criterion for successful underwriting (i.e., that a desk place 100% of underwritten securities into the market and avoid retaining a residual position) naturally helps preclude an underwriting desk from engaging in proprietary trading.

This post explains (a) underwriting RENTD and RENTD-based limits, (b) considerations for setting those limits, and (c) our view of where banks are in their implementation efforts and where they should be.

What is underwriting RENTD?

Underwriting RENTD is an estimate of market demand for the securities of an underwritten deal based on historical (and current) deal and market information. Factors used to calculate underwriting RENTD are generally the same factors that have been considered by underwriters to decide whether to take on an underwriting commitment and how to price it, including:

• Market liquidity, depth, and maturity
• Transaction pipeline for similar deal or asset types
• Previous deals and similar offerings
• Historic and current trends in deal participation
• Book building activities (e.g., marketing)
• Clients’ investment appetites

Therefore, underwriting RENTD can be estimated largely based on these factors, rather than by conducting a painstaking trade-by-trade customer demand assessment (as is required under the market making exemption). Further simplifying the analysis, there is effectively no restriction on who an underwriting desk may consider to be a “customer” (because the term “customer” is broadly defined to include all “market participants,” unlike the more restrictive definition under market making RENTD). [4]

It is noteworthy that prudent underwriters only agree to take on a deal to the extent that they can promptly place nearly 100% of the securities into the market. Therefore, in most cases, deals are oversubscribed over the notional amount in order to facilitate successful secondary trading after deal pricing. Following the same logic, RENTD can also be set at a range that exceeds 100% of a deal’s notional value.

Underwriting RENTD-based limits

Once a RENTD range is estimated, it should be augmented to take into consideration additional risks (e.g., reputational, operational, and valuation modelling) to arrive at the following three categories of limits:

• Size of underwriting positions
• Holding periods for residual securities
• Risks related to underwriting support activities

All desks taking the underwriting exemption must set these limits. However only “hard” underwriting (i.e., where a desk is actually committed to holding an underwritten position) counts toward a limit, while “best efforts” or “soft” underwriting (i.e., marketing and sales without holding a position) does not. Therefore, desks that intend to engage solely in soft underwriting have been considering effectively foregoing RENTD estimation and setting its three types of limits to zero. We do not recommend this approach, as discussed further below.

Importantly, the number of limits set for each desk within each of the three categories will vary based on the desk’s unique underwriting activities. As examples, a desk may establish only one limit in each of the three categories, or multiple limits under some categories based on the common characteristics of the various deals that the desk underwrites (the limits may differ, e.g., based on the issuer’s industry, credit quality, or jurisdiction of the deals).

Once these limits are set, they should be re-calibrated from time to time; however, they need not be re-calibrated on a deal-by-deal basis like RENTD (which changes every time a desk underwrites a new deal). A quarterly re-calibration of these limits is sensible.

Setting underwriting RENTD-based limits

Of the three areas in which RENTD-based limits are set, the size of the underwriting position is the one that benefits the most from existing policies and procedures. Most desks already calculate quantitative thresholds that are essentially the same as RENTD-based underwriting position limits and need very limited adaptation work to “Volckerize” these existing processes and thresholds to meet the Rule’s requirements.

Setting limits in the other two areas may seem more challenging, in part due to existing industry confusion around the Rule’s requirements, but is in fact very manageable as discussed below.

Holding periods

Although most underwriting desks aim to place 100% of underwritten securities into the market as part of the primary offering, it is common for an underwriting desk to retain some residual position. The Rule does not prohibit this holding of residuals, as forcing a trading desk to liquidate an issuer’s unsold securities onto the market would be detrimental to the issuer and to investors. However, the Rule does require a desk to limit the period of time that it holds underwritten securities in order to reduce the risk associated with holding an unsold allotment. [5]

Setting this limit presents some challenge, as some of the underlying factors necessary to estimate the appropriate holding time for residual positions are hard to predict (e.g., market liquidity). Therefore, the key to Volcker compliance in this area is to show that a trading desk’s residual underwriting position is under daily review and that controls are in place to help ensure that desks are making reasonable efforts to timely eliminate the residual risk.

A primary trading desk may also move a residual position to a secondary trading desk that is a market maker. The Rule does not to restrict this activity, as long as (a) the market making desk can show that it is ready to sell the residual position as a market maker and (b) the residual position becomes part of the desk’s market making inventory. However, banks should restrain the size of residual positions that underwriting desks take on so that market making desks are not stuck with large residual positions that are difficult to unwind.

Risks related to underwriting support activities

Although the Rule allows desks to engage in certain activities that support an underwritten deal, it requires that desks set limits to mitigate the risks arising from these activities. [6] These activities most notably include the following two:

• Stabilization—Where an underwriter purchases its issued securities in order to prevent the price from falling far below its issue price
• Issuer risk mitigation—Where a desk hedges risks on behalf of (and for) an issuer, such as FX risk arising from underwriting securities in a currency other than the issuer’s reporting currency (e.g., a foreign corporation issuing dollar-denominated securities)

Similar to the other two categories of limits, underwriters for the most part already consider the need to perform these supporting actions (and quantify the associated risks) when deciding whether or not to underwrite a deal in the first place. Therefore, this pre-deal analysis can contribute heavily toward calculating the limits for these activities.

It is noteworthy that some activities that have been historically performed to support underwriting are impermissible under the underwriting exemption (but rather should be taken within another exemption):

• Purchasing a financial instrument from a customer to help the customer buy securities in a new issuance
• Performing price discovery for underwritten securities by purchasing other similar financial instruments
• Hedging the trading desk’s underwriting position (including any residual positions) [7]

Other considerations

Treatment of securitizations under the underwriting exemption has been a point of confusion for the industry. Although the Rule allows a bank to accumulate assets or securities in anticipation of securitization or re-securitization (e.g., formation of an investment vehicle to issue collateralized loan obligations), this activity is not allowed under the underwriting exemption for financial instruments. Therefore, desks that are accumulating securities in anticipation of a re-securitization must do so under a different Volcker exemption. Accordingly, we are seeing banks often use the exemptions for permitted government and municipal obligations, or for market making (e.g., for non-agency mortgage backed securities related to re-securitization).

A second point of confusion has resulted from a subset of secondary transactions (i.e., block trades) that involve the filing of a registration statement or special selling efforts (e.g., marketing of block trades) above and beyond normal market making activities. In our view, these fall under the underwriting exemption instead of the market making exemption.

Where banks are and where they should be

While all banks are in the same boat in that they have yet to fully operationalize their underwriting desk-level compliance programs, closing the implementation gap for underwriting Volcker compliance should be one of the few relatively easy wins for banks.

Our observations suggest that nearly all banks have identified trading desks that will be taking the underwriting exemption and have policies and procedures in place that can be leveraged to substantially satisfy the Rule’s RENTD and limit setting requirements. As next steps, banks should assess how much their existing limits need to be tailored to meet Volcker requirements and what new limits, if any, should be added.

How to set and report underwriting limits is still an open question. With respect to limit setting, although certain underwriting desks may intend to forego RENTD estimation and set their limits to zero based on their restrained activities, banks should nevertheless consider setting limits even for these desks. For example, some underwriting desks that rarely, if ever, engage in “hard” underwriting are considering setting their underwriting position limits at zero. While this may seem reasonable, it means that even the slightest hard underwriting activity would breach the limits, subjecting the activity to a review and approval process. This situation may impair a desk’s ability to commit to potential underwriting opportunities that require prompt action. Thus, we think it is a better idea for these desks to set non-zero limits based on the potential size of positions they are willing to take on.

With respect to reporting limits and limit utilization (i.e., how far a desk is from breaching a limit), [8] some banks are planning to only count the size of their residual positions toward their underwriting position limit utilization, excluding the rest of the desks’ position commitments. Although this practice would lower limit utilization, we believe that banks should instead consider a desk’s position limit throughout the life of the deal (i.e., both before and after the deal is priced and securities are made available for trading). Therefore, in addition to post-deal residual positions, banks should also count their entire pre-deal underwriting position commitment toward their position limit.

Overall, banks should be able to leverage existing processes for setting and monitoring a trading desk’s underwriting limits substantially toward meeting the Rule’s underwriting exemption requirements. The key is to highlight existing underwriting processes and “Volckerize” them to make them part of the compliance program.

Endnotes:

[1] See PwC’s Regulatory brief, Volcker Rule: Are you really market making? (February 2015).
(go back)

[2] A residual position is the portion of underwritten securities that an underwriter does not sell as part of the primary offering (e.g., due to lack of demand), holding it instead until the securities can be sold at a reasonable price.
(go back)

[3] These limits must be in place by July 21st, as part of the standard compliance program. See PwC’s A closer look, Volcker rule clarity: Waiting for Godot (May 2014).
(go back)

[4] This broad definition of customers reflects the fact that the dealer-to-dealer trading activity that is prevalent in secondary trading among market making desks is irrelevant during a primary deal, as underwritten securities are almost always placed with buy-side customers.
(go back)

[5] Most banks on their own also restrict the amount of a deal’s securities that can be held as residual. However, the Rule addresses unsold allotment risk primarily by limiting the holding period, as other aspects of the risk are already captured in setting the underwriting position size limit.
(go back)

[6] Setting these limits is only necessary when support activities are carried out by the underwriting desk itself, rather than by secondary desks (e.g., market making desks).
(go back)

[7] There is no permissible hedging of a desk’s risk under the underwriting exemption, in contrast to the market making exemption which permits certain hedging. Therefore, desks that hedge underwritten positions must do so utilizing the Rule’s risk mitigating hedging exemption.
(go back)

[8] Limits and limit utilization are reportable as of July 21, 2015 for the nine largest trading banks, and in 2016 for smaller banks.
(go back)