Pub. 7 2016 Issue 1

www.wvbankers.org 28 West Virginia Banker T he risk-adjusted return on any asset should be conducive to its risk. If a bank makes two loans, it prices each loan based on the risk of the underlying borrower. If one of the borrowers has high- er risk, the loan needs to provide a higher return. This is fundamental to banking, and high performing banks price risk effectively. Banks typically adhere to this practice in lending but often are less diligent when managing the bond portfolio. When a bank buys a bond, it inherently accepts risk for a return, the same as when it makes a loan. Bonds have risk (call, extension, liquidity, duration, etc.), and banks need to be rewarded adequately. West Virginia banks have over 20% of their assets in the bond portfolio, so it is a substantial component of return and risk. One of the best metrics a bank can use to gauge the “Real Return” of the bond portfolio is to divide the yield (return) by the effective duration (risk). Effective duration is a price sensitivity measure, and it measures this price sensitivity for every 100 bps of interest rate change. If a bond has a 4.00% effective duration, it loses 4% of its value for every +100bps of rate increase (i.e. +300bps = -12% price volatility). Effective duration is a stronger risk measurement tool than modified duration or average life because it incorporates embedded options that are routinely in the bonds (Callable Agencies/Munis, MBS, CMO’s, etc.) that banks hold. If a bond is likely to be called, extend, or be prepaid, effective duration captures that risk. Hence, the yield/duration ratio provides a reward-to-risk measurement that can be used as a buying tool for new bonds as well as a “report card” for the current portfolio. If a bond portfolio can maintain a yield/duration ratio of 75% or greater (Table 1), the portfolio is generating a good return-to-risk. If the yield/duration of the portfolio is below 50%, the portfolio has a low return given its risk level. Reward to Risk Ratio (Yield /Eff. Duration) Good Performance (Yld/Dur) > 75% Average Performance (Yld/Dur) 50% - 75% Assessing the “Real Return” of the Bond Portfolio Disclaimer: This material has been prepared by Duncan-Williams, Inc., member FINRA/SIPC (Duncan-Williams”), from information sources believed to be reliable. Duncan-Williams expressly disclaims any and all liability which may be based on such information, errors therein or omissions there from, or in any other written or oral communication related thereto. This material is for your information only and should not be construed as investment advice or as any recommendation of a transaction. Neither Duncan-Williams nor any of its representatives is soliciting any action based upon this material. Specifically, this material is not, and is not to be construed as, an offer to buy or sell any security or other financial instru- ment referred to herein. Take a look at Portfolios 1 and 2 in Table 2. From a yield standpoint, these portfolios look identical. They both return 2.00%; however, stopping at yield alone would fail to capture the real performance. Portfolio 2 has a much higher effective duration of 4.18%, compared to 2.60% in Portfolio 2. Portfolio 1 has a high performing reward-to-risk ratio of 77% (2.00% / 2.60%) compared to Portfolio 2 that has an underperforming ratio of 48% (2.00% / 4.18%). Even though both portfolios yield the same, the interest rate risk, EVE, and capital exposure is greater for Portfolio 2. Therefore, Portfolio 1 is a much higher performing portfolio. Table 1 Table 2 Market Value Yield Effective Duration Yield / Effective Duration % Change +300 $ Loss +300 Portfolio 1 $50,000,000 2.00% 2.60% 77% -7.8% ($3,900,000) Portfolio 2 $50,000,000 2.00% 4.18% 48% -12.5% ($6,270,000) In Table 3, we compiled results for around 10,000 bonds which were analyzed during the fourth quarter of 2015 and held by banks from around the country. We can see the yield and duration for each sector and the return/risk ratio. At the lower range are Agency bullets at 34% (1.40%/4.14%) with ARMs at the higher end at 126% (1.22%/0.97%). Agency bullets have four times (4.14%) the price risk as ARMs (0.97%), but the yields are relatively close (1.40% vs. 1.22%). If we assumed a $1mm investment in both of these sectors and an immediate shock of +300bps, the Agency bullet would lose $124,200 (4.14% x $1mm x 3) while the ARMwould only lose $29,100 (0.97% x $1mm x 3). That is high price risk for minimal pick-up in yield. Table 3 This metric works very well in conjunction with a bank’s asset and liability needs. For example, if a bank is liability-sensitive and trying to hedge against higher interest rates, they could overweight the portfolio with Hybrid ARMs, SBAs or ARMs. These three categories add defense with an attractive yield. Likewise, if a bank is asset-sensitive or less concerned with higher interest rates, they could overweight tax-free Municipals and Corporates for more return. If a bank did not want the credit risk from those two sectors, then they move down to Fixed MBS and CMOs. Most portfolio accounting or portfolio analytic systems provide book yield and effective duration. Using this simple ratio provides a great barometer of how well the portfolio is performing on a risk-adjusted basis.