Progressive 2014 Annual Report Download - page 24

At December 31, 2014, vendor-quoted prices comprised 97% of our Level 2 classifications (excluding short-term
investments), while dealer-quoted prices represented 3%, compared to 98% and 2% at December 31, 2013, respectively. In
our process for selecting a source (e.g., dealer, pricing service) to provide pricing for securities in our portfolio, we reviewed
documentation from the sources that detailed the pricing techniques and methodologies used by these sources and
determined if their policies adequately considered market activity, either based on specific transactions for the particular
security type or based on modeling of securities with similar credit quality, duration, yield, and structure that were recently
transacted. Once a source is chosen, we continue to monitor any changes or modifications to their processes by reviewing
their documentation on internal controls for pricing and market reviews. We review quality control measures of our sources
as they become available to determine if any significant changes have occurred from period to period that might indicate
issues or concerns regarding their evaluation or market coverage.
As part of our pricing procedures, we obtain quotes from more than one source to help us fully evaluate the market price of
securities. However, our internal pricing policy is to use a consistent source for individual securities in order to maintain the
integrity of our valuation process. Quotes obtained from the sources are not considered binding offers to transact. Under our
policy, when a review of the valuation received from our selected source appears to be outside of what is considered market
level activity (which is defined as trading at spreads or yields significantly different than those of comparable securities or
outside the general sector level movement without a reasonable explanation), we may use an alternate source’s price. To
the extent we determine that it may be prudent to substitute one source’s price for another, we will contact the initial source
to obtain an understanding of the factors that may be contributing to the significant price variance, which often leads the
source to adjust their pricing input data for future pricing.
To allow us to determine if our initial source is providing a price that is outside of a reasonable range, we review our
portfolio pricing on a weekly basis. We frequently challenge prices from our sources when a price provided does not match
our expectations based on our evaluation of market trends and activity. Initially, we perform a review of our portfolio by
sector to identify securities whose prices appear outside of a reasonable range. We then perform a more detailed review of
fair values for securities disclosed as Level 2. We review dealer bids and quotes for these and/or similar securities to
determine the market level context for our valuations. We then evaluate inputs relevant for each class of securities
disclosed in the preceding hierarchy tables.
For our structured debt securities, including commercial, residential, and asset-backed securities, we evaluate available
market-related data for these and similar securities related to collateral, delinquencies, and defaults for historical trends and
reasonably estimable projections, as well as historical prepayment rates and current prepayment assumptions and cash
flow estimates. We further stratify each class of our structured debt securities into more finite sectors (e.g., planned
amortization class, first pay, second pay, senior, subordinated, etc.) and use duration, credit quality, and coupon to
determine the appropriate fair value.
For our corporate debt and preferred stock (redeemable and nonredeemable) portfolios, we review securities by duration,
coupon, and credit quality, as well as changes in interest rate and credit spread movements within that stratification. The
review also includes recent trades, including: volume traded at various levels that establish a market, issuer specific
fundamentals, and industry specific economic news as it comes to light.
For our municipal securities (e.g., general obligations, revenue, and housing), we stratify the portfolio to evaluate securities
by type, coupon, credit quality, and duration to review price changes relative to credit spread and interest rate changes.
Additionally, we look to economic data as it relates to geographic location as an indication of price-to-call or maturity
predictors. For municipal housing securities, we look to changes in cash flow projections, both historical and reasonably
estimable projections, to understand yield changes and their effect on valuation.
Lastly, for our short-term securities, we look at acquisition price relative to the coupon or yield. Since our short-term
securities are typically 90 days or less to maturity, with the majority listed in Level 2 being seven days or less to redemption,
acquisition price is the best estimate of fair value.
We also review data assumptions as supplied by our sources to determine if that data is relevant to current market
conditions. In addition, we independently review each sector for transaction volumes, new issuances, and changes in
spreads, as well as the overall movement of interest rates along the yield curve to determine if sufficient activity and liquidity
exists to provide a credible source for our market valuations.
During each valuation period, we create internal estimations of portfolio valuation (performance returns), based on current
market-related activity (i.e., interest rate and credit spread movements and other credit-related factors) within each major
App.-A-23