Unum 2015 Annual Report Download - page 109

107
Unum 2015 Annual Report
The parameters and inputs used to validate a price on a security may be adjusted for assumptions about risk and current market
conditions on a quarter to quarter basis, as certain features may be more significant drivers of valuation at the time of pricing. Changes to
inputs in valuations are not changes to valuation methodologies; rather, the inputs are modified to reflect direct or indirect impacts on asset
classes from changes in market conditions.
Fair values for derivatives other than embedded derivatives in modified coinsurance arrangements are based on market quotes or
pricing models and represent the net amount of cash we would have paid or received if the contracts had been settled or closed as of the
last day of the period. We analyze credit default swap spreads relative to the average credit spread embedded within the LIBOR-setting
syndicate in determining the effect of credit risk on our derivatives’ fair values. If net counterparty credit risk for a derivative asset is
determined to be material and is not adequately reflected in the LIBOR-based fair value obtained from our pricing sources, we adjust the
valuations obtained from our pricing sources. For purposes of valuing net counterparty risk, we measure the fair value of a group of financial
assets and financial liabilities on the basis of the price that would be received to sell a net long position or transfer a net short position for a
particular risk exposure in an orderly transaction between market participants at the measurement date under current market conditions. In
regard to our own credit risk component, we adjust the valuation of derivative liabilities wherein the counterparty is exposed to our credit
risk when the LIBOR-based valuation of our derivatives obtained from pricing sources does not effectively include an adequate credit
component for our own credit risk.
Fair values for our embedded derivative in a modified coinsurance arrangement are estimated using internal pricing models and
represent the hypothetical value of the duration mismatch of assets and liabilities, interest rate risk, and third party credit risk embedded in
the modified coinsurance arrangement.
Certain of our investments do not have readily determinable market prices and/or observable inputs or may at times be affected by
the lack of market liquidity. For these securities, we use internally prepared valuations combining matrix pricing with vendor purchased
software programs, including valuations based on estimates of future profitability, to estimate the fair value. Additionally, we may obtain
prices from independent third-party brokers to aid in establishing valuations for certain of these securities. Key assumptions used by us to
determine fair value for these securities include risk-free interest rates, risk premiums, performance of underlying collateral (if any), and
other factors involving significant assumptions which may or may not reflect those of an active market.
At December 31, 2015, approximately 6.8 percent of our fixed maturity securities were valued using active trades from TRACE pricing
or broker market maker prices for which there was current market activity in that specific security (comparable to receiving one binding
quote). The prices obtained were not adjusted, and the assets were classified as Level 1, the highest category of the three-level fair value
hierarchy classification wherein inputs are unadjusted and represent quoted prices in active markets for identical assets or liabilities.
The remaining 93.2 percent of our fixed maturity securities were valued based on non-binding quotes or other observable and
unobservable inputs, as discussed below.
• Approximately 78.3 percent of our fixed maturity securities were valued based on prices from pricing services that generally use
observable inputs such as prices for securities or comparable securities in active markets in their valuation techniques. These assets
were classified as Level 2. Level 2 assets or liabilities are those valued using inputs (other than prices included in Level 1) that are
either directly or indirectly observable for the asset or liability through correlation with market data at the measurement date and
for the duration of the instrument’s anticipated life.
• Approximately 3.4 percent of our fixed maturity securities were valued based on one or more non-binding broker quotes, if validated
by observable market data, or on TRACE prices for identical or similar assets absent current market activity. When only one price is
available, it is used if observable inputs and analysis confirms that it is appropriate. These assets, for which we were able to validate
the price using other observable market data, were classified as Level 2.
• Approximately 11.5 percent of our fixed maturity securities were valued based on prices of comparable securities, matrix pricing,
market models, and/or internal models or were valued based on non-binding quotes with no other observable market data. These
assets were classified as either Level 2 or Level 3, with the categorization dependent on whether there was other observable market
data. Level 3 is the lowest category of the fair value hierarchy and reflects the judgment of management regarding what market
participants would use in pricing assets or liabilities at the measurement date. Financial assets and liabilities categorized as Level 3
are generally those that are valued using unobservable inputs to extrapolate an estimated fair value.