0001144204-11-048084.txt : 20120314 0001144204-11-048084.hdr.sgml : 20120314 20110817095720 ACCESSION NUMBER: 0001144204-11-048084 CONFORMED SUBMISSION TYPE: CORRESP PUBLIC DOCUMENT COUNT: 2 FILED AS OF DATE: 20110817 FILER: COMPANY DATA: COMPANY CONFORMED NAME: FIDELITY D & D BANCORP INC CENTRAL INDEX KEY: 0001098151 STANDARD INDUSTRIAL CLASSIFICATION: NATIONAL COMMERCIAL BANKS [6021] IRS NUMBER: 233017653 STATE OF INCORPORATION: PA FISCAL YEAR END: 1231 FILING VALUES: FORM TYPE: CORRESP BUSINESS ADDRESS: STREET 1: BLAKELY & DRINKER STREETS CITY: DUNMORE STATE: PA ZIP: 18512 BUSINESS PHONE: 5703428281 MAIL ADDRESS: STREET 1: BLAKELY & DRINKER STREETS CITY: DUNMORE STATE: PA ZIP: 18512 CORRESP 1 filename1.htm Unassociated Document
www.bankatfidelity.com
1 (800) 388-4380
 
 
 
August 17, 2011
 
 
 
Mark Webb
Legal Branch Chief
United States Securities and Exchange Commission
Washington, D.C. 20549

 
Re:
Fidelity D & D Bancorp, Inc.
 
Form 10-K for Fiscal Year Ended
 
December 31, 2010
 
Filed March 29, 2011
 
Forms 10-Q for Fiscal Quarter Ended
 
March 31, 2011
 
Filed May 11, 2011
 
File No. 333-90273

Dear Mr. Webb:

In response to the staff’s comment letter dated August 4, 2011 concerning the above referenced filings for Fidelity D & D Bancorp, Inc. (the “Company”), the requested information is submitted below. The numbered paragraphs set forth below correspond to the numbered paragraphs in the staff’s comment letter.

Form 10-K for Fiscal Year Ended December 31, 2010
 
Business, page 3
 
1.  
We note from page 4 that you are significantly dependent on local economic conditions. In future filings, please describe the principal economic features and condition of your market area as relevant to your company. As warranted, consider including figures for change in average income, home sales and prices, unemployment, and the like.

Management’s Response:

In future filings, the Company will include a description of the principal economic features and condition of the Company’s market area.
 
 
 
 
Blakely & Drinker Streets
Dunmore, PA 18512
Tel: (570) 342-8281
Fax: (570) 356-5724
 
 
 
Mark Webb
August 17, 2011
Page 2
 
Form 10-Q for Fiscal Quarter Ended March 31, 2011
 
Financial Statements
 
Note 7 – Fair Value Measurements, page 22
 
2.  
We note your tabular presentation on page 25 reporting that impaired loans and other real estate owned use Level 2 and Level 3 inputs to determine their fair value.  Please note that due to the fact the appraisal process tends to be very assumption driven with comparatives, expectations on cash flows, and other factors, we believe that even though an appraisal is current and performed by an independent third party, it still would not appear to meet the Level 2 reporting for fair value.  Please tell us why you believe the fair value of impaired loans and other real estate owned that are measured using current appraisals are not all considered to use Level 3 inputs.
 
Management’s Response:

The Company utilized current appraisals based upon comparable sales to derive the Level 2 fair value of impaired loans and other real estate owned.  We believed the comparable sales data to be market observable inputs that did not require significant adjustments to arrive at a fair value.  We also recognize the fact that the appraisal process tends to be very assumptions driven based on factors such as comparables and expectations on cash flows. Therefore, we concur that even though an appraisal is current and performed by an independent third party, it may still not meet the Level 2 reporting standard for fair value.  As such, as of June 30, 2011 and forward, we will classify impaired loans and other real estate owned supported by an appraisal to be a Level 3 fair value.

Management’s Discussion and Analysis, page 25
 
Investment Securities, page 30
 
3.  
Based on our review of your disclosure, it appears you are utilizing the guidance in ASC 325-40-15 and 35 in your other than temporary impairment (“OTTI”) methodology.  If so, please tell us why you believe the trust preferred securities are within the scope of this Guidance. Additionally, please tell us the credit ratings of the securities at the time of acquisition.
 
Management’s Response:

The pooled trust preferred collateralized debt obligations (“TruPS CDO”) are beneficial interests in securitized financial assets, which as of the 2008 credit market crisis onset, were within the scope of Emerging Issues Task Force Issue No. 99-20, Recognition of Interest Income and Impairment on Purchased Beneficial Interests and Beneficial Interests That Continue to Be Held by a Transferor in Securitized Financial Assets.  The Company’s portfolio of TruPS CDO securities are categorized as bank trust preferred securities, which were issued by pooling borrowings of small to midsize bank holding companies.  After being pooled, the borrowings were securitized as financial assets structured in equity form.  The resulting tranches possess debt instrument characteristics that meet the beneficial interest characteristics under ASC 325-40-15-3 guidance.  The Company periodically purchased the securities from 2002 through 2008.  Two tranches that were initially rated “single-A”, at the time of acquisition by nationally recognized statistical rating organizations, had been written down to zero book value from credit related OTTI charges realized through September 30, 2010.  At December 31, 2010, the portfolio consisted of 11 tranches, nine of which were initially rated “single-A” and two of which were initially rated “double-A” at the time of acquisition by nationally recognized statistical rating organizations.  The portfolio has provided cash flows in the form of coupon payments and principal pay downs.

 
 
Mark Webb
August 17, 2011
Page 3
 
Initially working with FTN Financial Capital Markets’ Structured Finance Group, the Company relied on:

1)  
A stress test analysis to determine the sufficiency of the collateral subordination to ensure that the possibility of credit loss (whether with respect to principal or interest) is remote, and

2)  
Due to the lack of guidance for securities whose cash flows change as a result of both prepayments and credit losses, the provisions of the FASB Accounting Standards Codification (FASB ASC) 325-40-35, Beneficial Interests in Securitized Financial Assets, set forth the rules for determining when securitized financial asset securities must be written down to determine fair value because of impairment.  The impairment provisions of FASB ASC 325-40-35 brought us much closer to a lower-of-cost or fair value approach.  If fair value is less than amortized cost and the present value of the estimated cash flows have significantly decreased, resulting in a probable adverse change since the last estimate was made (other than as a result of just an interest rate reset), then the security was written-down to the determined fair value through earnings. 

This guidance also retains and emphasizes OTTI assessment and the related disclosure requirements in FASB ASC 320 accounting for Investments -Debt and Equity Securities and other related guidance. The changes align the OTTI methodology for certain securitizations (such as TruPS CDO transactions) with the approach used in accounting for Investments - Debt and Equity Securities.  By moving to this approach, investments in debt and equity securities accounting did not require exclusive reliance on market participant assumptions regarding future cash flows, and permitted the use of reasonable management judgment of the probability that the holder will be unable to collect all amounts (principal and interest) due, which could result in OTTI.  We were then able to focus more on our judgment of whether it is probable that we will collect all amounts due. 

 
 
Mark Webb
August 17, 2011
Page 4
 
4.  
Please tell us if the discount rate used represents the current yield used to accrete the beneficial interest, the effective interest rate implicit in the security at the date of acquisition, or something else.  Please provide us the accounting guidance that supports your discount rate.
 

Management’s Response:

The discount rates utilized as of December 31, 2010 and March 31, 2011 were calculated by a discount rate adjustment technique pursuant to ASC 820-10-55-10 derived from observed rates of return for comparable floating rate assets (Bloomberg U.S. Bank unsecured debt A and/or BBB index) interpolated to the credit rating level based upon the underlying portfolio credit risk and the security’s weighted average life.  The factors were utilized as they represent observable market inputs and they were adjusted for credit risk considerations embedded within individual obligor credit-worthiness estimates.

 
5.  
Please provide us with a listing of each trust preferred security as of December 31, 2010 and March 31, 2011, with the actual discount rate used, the current yield used to accrete the beneficial interest, and the effective interest rate implicit in the security at the date of acquisition. Please explain to us the financial impact on expected cash flows and credit OTTI, if you had used the effective interest rate implicit in the security at the date of acquisition as the discount rate.
 
Management’s Response:

The follow table lists each of the aforementioned 11 trust preferred securities held by the Company.  Due to the fact that each security pays a coupon based on a floating rate tied to three-month LIBOR, a discount margin methodology is an appropriate metric to price each security.  That is, the discount margins are added to the three-month forward LIBOR to discount the projected CDO tranche cash flows. Three different margins are provided below: the margin implicit at the date of acquisition, discount margin used to accrete the beneficial interest and the actual discount margin used for calculating fair value.  The latter two discount margins are provided as of December 31, 2010and March 31, 2011 and were applied to CDO tranche cash flows generated under Monte Carlo simulation. For avoidance of doubt, the CDO tranche cash flows reflect projected losses due to default on the underlying collateral:

 
 
Mark Webb
August 17, 2011
Page 5
 
Spread to forward 3 month LIBOR
December 31, 2010
March 31, 2011
Deal
Name
Discount Margin implicit at date of acquisition
Discount margin used to accrete beneficial interest
Discount Margin for actual discount rate used
Discount margin used to accrete beneficial interest
Discount Margin for actual discount rate used
4
PRETSL IV Mezz
2.198%
2.186%
2.390%
2.186%
1.350%
             
5
PRETSL V Mezz
2.162%
2.100%
N/A
2.100%
N/A
             
7
PRETSL VII B
1.800%
1.800%
2.680%
1.800%
N/A
             
9
PRETSL IX B1
1.718%
1.800%
3.100%
1.800%
3.410%
9
PRETSL IX B3
1.876%
1.800%
3.100%
1.800%
3.410%
             
11
PRETSL XI B3
1.990%
1.600%
3.680%
1.600%
3.810%
             
15
PRETSL XV B1
1.570%
1.570%
3.250%
1.570%
N/A
             
16
PRETSL XVI C
1.379%
1.550%
N/A
1.550%
N/A
             
17
PRETSL XVII C
1.260%
1.400%
3.620%
1.400%
N/A
             
18
PRETSL XVIII C
1.350%
1.350%
3.700%
1.350%
3.130%
             
19
PRETSL XIX C
1.148%
1.200%
3.950%
1.200%
3.230%
             
24
PRETSL XXIV B1
1.529%
0.580%
4.290%
0.580%
3.560%
             
25
PRETSL XXV C1
0.850%
0.850%
3.250%
0.850%
N/A
             
27
PRETSL XXVII B
0.951%
0.600%
3.430%
0.600%
2.830%
 
N/A = The discount margin is not applicable due to the fact the security's amortized cost was previously written down to zero from credit related OTTI charges realized.
The impact of converting to a different valuation methodology, as of December 31, 2010, was higher default rates and higher pool losses.  This was a result of reviewing the credit worthiness of each underlying obligor for each transaction.  It was also a result of adopting a more robust methodology that took into account a fairly conservative estimate with respect to default correlation between any obligors (50%). Lastly, the methodology adopted as of December 31, 2010 assumes zero recovery with respect to obligor defaults.  It is estimated that roughly 85-90% of the portfolio book value decrease was due to changes in the default and recovery methodologies and the remaining 10-15% of the decrease in portfolio book value was due to a change in the discount rate methodology.

 
 
Mark Webb
August 17, 2011
Page 6
 
Please refer to Management’s Response to Comment 7 for a further explanation of the financial impact and credit OTTI differences as of December 31, 2010.

6.  
Please provide us with your analysis of the credit portion of the OTTI that would have been recognized if you had used the impairment model in ASC 320-10-35-17 and please explain the primary factors that you believe caused any differences.

Management’s Response:

In response to this comment, Bybel Rutledge LLP, counsel to the Company, will furnish to the staff copies of the proprietary reports provided to the Company by a third party vendor. These reports are being provided under separate cover requesting confidential treatment pursuant to the provisions of C.F.R. Sec. 200.83 and Rule 12b-4 promulgated under the Securities Exchange Act of 1934, as amended. In accordance with such rule, counsel to the Company will request that those materials be returned promptly following the completion of the staff’s review thereof.

7.  
We note your disclosure stating that prior to December 31, 2010 you used a different valuation process for determining whether there was OTTI on your trust preferred securities. Please see our comments below:
 
·    
Tell us your primary reasons for changing your valuation process;
·    
Provide us an analysis that explains the primary differences that the new model yielded;
·    
Tell us why you believe the new model resulted in $12 million in credit losses as compared to the old model;
·    
Tell us the specific assumptions used in the new model that you believe had the greatest impact on the current results;
·    
For the assumptions that had the most significant changes under the new model, tell us the assumption, the amount used at September 30, 2010 and the amount used at December 31, 2010, explain why you revised the assumption and quantify the impact on credit OTTI from the change

Management’s Response:

As was noted in the 2nd paragraph of Management’s Response to Comment 5 above, the Company moved to a different valuation methodology as of December 31, 2010.  The primary reason was to move away from the use of a single credit scenario assumption with respect to the global default rate of all underlying collateral within a TruPS CDO transaction. We moved towards a more robust methodology that relies on the individual default probabilities of each underlying collateral obligor and an assumption about default correlation between obligors.

 
 
Mark Webb
August 17, 2011
Page 7
 
Prior to December 31, 2010, the valuation method used was different than the process currently used.  The inputs used in the past consisted of a mix of both observable and unobservable criteria. The inputs resulted in a single default curve that was applied to calculate the expected CDO tranche cash flow under one scenario. For example, prior to December 31, 2010, to project a default rate, universal adjustments were applied to the historical average of default rates of banks.  The historical average default rates were obtained from the FDIC for U.S. Banks and Thrifts for the period spanning 1988 to 1991.  This rate was tripled and the results were then stratified beginning with a higher rate of default and then regressing to normal, with projected global recoveries and prepayment speeds. The below table contains the applied default curve, broken down by period:

Time Period
Default Rate (not annualized)
Recovery Rate
Q2 2009
6.00%
15% with 2 Yr Lag
Q3 2009
4.00%
15% with 2 Yr Lag
Q4 2009
3.00%
15% with 2 Yr Lag
Q1 2010
2.00%
15% with 2 Yr Lag
Q2 2010
1.00%
15% with 2 Yr Lag
Q3 2010
1.00%
15% with 2 Yr Lag
Q4 2010
1.00%
15% with 2 Yr Lag
Thereafter
0.25%
15% with 2 Yr Lag

The curve was applied and adjusted each quarter for the actual amount of to-date deferrals/defaults in all TruPS CDO transactions.  For example, if a given quarter had a 2.0% projected default rate but actually experienced a 2.5% default rate, the actual rate replaced the projected default rate.  Thus, projected defaults were replaced with actual results once the given quarter had transpired.  The resulting projected default rate was then applied using Intex to calculate tranche cash flows for each transaction in the Company’s portfolio.  A discount rate, equal to the three-month LIBOR forward rate plus the margin implicit at the date of acquisition, was used to calculate a fair value price. This was done to determine the then current period-end valuations and the existence of OTTI.

 
 
Mark Webb
August 17, 2011
Page 8
 
This set of global default assumptions ran its course in September 2010 and new default assumptions were required after the fourth quarter of 2010.  Both company management and the bank’s Audit Committee were uncomfortable with simply switching to another set of global default assumptions, as we felt the methodology could no longer be adequately substantiated.  With the Board of Director’s approval, management engaged Black Swan Consulting ("Black Swan"), a Structured Products Specialist Firm, to (i) propose methodologies with respect to calculating Fair Value, Other Than Temporary Impairment and Credit Loss; (ii) propose model inputs including individual obligor default probabilities, recovery rates, base correlations and discount rates; (iii) run Monte Carlo simulations and provide sensitivity analysis with respect to model inputs; and (iv) and provide the results in report format.

Black Swan conducted a review of each of the twelve and then eight pooled trust preferred securities with remaining book values as of December 31, 2010 and March 31, 2011, respectively.  Black Swan accomplished this by performing the following for each TruPS CDO transaction:

1)  
Reviewed each underlying piece of collateral and assigned a default probability based on publicly available data with respect to obligor credit-worthiness.
2)  
Reverse-engineered each TruPS CDO transaction by creating a cash model that projected the expected CDO tranche cash flows based on the characteristics of the underlying collateral and the waterfall structure of the CDO.
3)  
Projected time-to-default for each piece of collateral under 10,000 Monte Carlo simulation trials based on the individual default probabilities assigned and an assumption about correlation with respect to default.
4)  
Simulated 10,000 Monte Carlo time-to-default scenarios through the cash flow model to determine the expected CDO tranche cash flow as well as  to measure the percentage of scenarios that resulted in a loss to the tranche
5)  
Utilized observable market data with several factors to construct an applicable discount rate for each CDO tranche:
a.    
The starting point was the portfolio's weighted average rating and weight average life, which was used to determine the closest USD US Bank Bloomberg Fair Value market curve discount rate to the actual portfolio.
b.    
An appropriate spread differential between the rating assumed by the Bloomberg US Bank index spread and the actual weighted average credit quality of the transaction’s collateral pool was then added.
c.    
Because the Bloomberg Fair Value market discount rate curves are calculated from no default cash flows and we are applying the constructed discount rate to projected tranche cash flow under Monte Carlo simulation that includes losses, the credit loss projected under Monte Carlo simulation was subtracted as to avoid double counting of credit risk.

6)  
Lastly, Black Swan calculated the appropriate fair value price for each of the securities, given the applicable discount rate.  In addition to calculating fair value, Black Swan measured the percentage of scenarios under simulation that projected principal loss to the CDO tranche.

 
 
Mark Webb
August 17, 2011
Page 9
 
The new valuation model resulted in $9.3 million of fourth quarter credit OTTI charges at December 31, 2010, which was realized.  The old model that utilized global default assumptions was also run for December 31, 2010.  When its estimated cash flows were discounted at the effective interest discount margin implicit in the security as of the date of acquisition, it resulted in approximately $2.1 million of fourth quarter credit OTTI charges.  As was noted above, the new model resulted in additional credit losses due primarily to the higher projected defaults.  The higher defaults were calculated from reviewing the credit quality of each piece of collateral in each of the collateral pools.  As a result, credit losses had the greatest impact on the portfolio valuation results.


In responding to your comments, the Company acknowledges that:
 
·    
the Company is responsible for the adequacy and accuracy of the disclosure in the filing;
 
·    
staff comments or changes to disclosure in response to staff comments do not foreclose the Commission from taking any action with respect to the filing; and
 
·    
the Company may not assert staff comments as a defense in any proceeding initiated by the Commission or any person under the federal securities laws of the United States.
 
If you have any questions, please contact either Erik Gerhard at (717) 731-1700 or the undersigned at (570) 504-8000.
 
 
Sincerely,
 
/s/ Salvatore R. DeFrancesco, Jr.

Salvatore R. DeFrancesco, Jr.
Chief Financial Officer
 
 
 
 
GRAPHIC 2 logo.jpg GRAPHIC begin 644 logo.jpg M_]C_X``02D9)1@`!`0$`8`!@``#_VP!#``@&!@<&!0@'!P<)"0@*#!0-#`L+ M#!D2$P\4'1H?'AT:'!P@)"XG("(L(QP<*#7J#A(6&AXB)BI*3E)66EYB9FJ*CI*6FIZBIJK*SM+6VM[BYNL+#Q,7& MQ\C)RM+3U-76U]C9VN'BX^3EYN?HZ>KQ\O/T]?;W^/GZ_\0`'P$``P$!`0$! M`0$!`0````````$"`P0%!@<("0H+_\0`M1$``@$"!`0#!`<%!`0``0)W``$" M`Q$$!2$Q!A)!40=A<1,B,H$(%$*1H;'!"2,S4O`58G+1"A8D-.$E\1<8&1HF M)R@I*C4V-S@Y.D-$149'2$E*4U155E=865IC9&5F9VAI:G-T=79W>'EZ@H.$ MA8:'B(F*DI.4E9:7F)F:HJ.DI::GJ*FJLK.TM;:WN+FZPL/$Q<;'R,G*TM/4 MU=;7V-G:XN/DY>;GZ.GJ\O/T]?;W^/GZ_]H`#`,!``(1`Q$`/P#W^BBB@`HH MHH`***;)(D2,\C*J*,LS'``]2:`'45ST_B5K@E-'MOM/8W,I*0CZ'J_X#'O6 M=/;3WP)U2]DN%/6%/W4(_P"`@Y;_`($308NLOLZFY=>)-*M9##]I$TX_Y8VZ MF5\^X7./QQ5"3Q'?3<6FE>6IZ/=S!?\`QU=Q_450$UI:Q^5`J*@Z)$H`_3BH M'OF)^1`/KS092J2?7[BW)?:W-R^HPVX_NVUN/YN6_E59XKF4YEU347/M<%!^ M28J!KB9NKD?3BHRS'JQ/U-!FWPMC_`*SS'/\`TTG=OYM41TW3O^?:$_49 MI**";1["?V;8#I;1#Z<5*EE`N/+:>/WCN)%_DU1T4!9=BXBW<7,.JZ@A[;IO M,'_CX-68]2UR`?+>VUR/2XM]I_-"/Y5EAF'1B/QIXN)1_%GZT%)VV-V+Q-=1 M?\?FDR%?[]I*)1_WR=I_+-:5EK^F7\GE0W:";_GC*#')_P!\M@URR7N/OI^1 MJ23['?)Y<\<BN-@2^L/\`D'7SH@_Y=[G,L?T!/S+^ M!_"M2V\31*ZQ:I";&1CA9"VZ%S[/V^C8H-557VM#>HI`UI[N4?NK:/EW]_9?4GBL&6VGU%O/UB59%4[EM4.(8_KG[Y]SQZ"G) M%%IJR7$\K3W4QS+.X^>4^@'91V`X%U!R5 M:B2O/;M_F;\^HHORPC<1QD\`?2L^6:28Y=R?;M7G]YX_N6)6RM(HE_O2DNWY M#`'ZUDOX@U^_)VWER?:W7;_Z"!0<,\;#9:GJ9&!DC`]:0.K=&4_0UY))INLW M1WRV.HS'^\T,C?J147]CZE_T#+S_`,!W_P`*#/ZW+^0]B`+=!GZ4NQO[I_*O M&)=.O((VDFLKB-%ZL\+`#ZDBJX(7T%!+QK6\?Q_X![<5(Z@BDKQ[37U`7:C3 M&G^T9X$)/Z^WUXKU^/?Y2>;@2;1OV],XYQ0=%"M[5/2PZBL'Q!XFCT*6&(6W MVB612Q7S-NT9P,\'KS^57]'O9]1TR.\G@6`RY*(&)^7L2?>@T52+ERIZE^BB M@G').!ZT%A17)1>-_M6HK9V>F&4R2;(W:?`(S]XC:<#O76T$0J1G\+)(KB6' M[K9;]6L96^7_MFW\!] MNGTKHM-U6VU2%G@+*\9VRPR#:\;>C#M_(]JY"VO7@PK9:/T[CZ5>D@,\D=]8 M3"&\0827'##^XX[K^HZB@TA-Q^';M_D==16?I.JIJ<+AHS#=0G;/`QR4;Z]U M/4'O6A0=49*2N@HHHH&%%%%`!1110!3U/4(],L)+J12^W`1%ZNQ.%4>Y)`KG MX4:Q@FO+U@][.=TS+Z]D7_9'0?B:LW[_`&_Q"L/6#3D$A'K,X^7_`+Y7)_X& M*RM3N3/ M@XN15'S3U*MOIEA:`"WLK>/'=8AG\^M.-\!J:V)5PS0&97SP0&`(^O(/XU8K M(GE63Q790QG+P6TKR8[!BH'ZB@M^ZE8UB0`68@`#))/`%<;K/CZP[7>H,,@3'^ M+K?BO4;3Q$]K&$-M"ZJ860'S.`3SU'7C'M777]_;:;:FYNY-D0(7."22?85G M6-QHNO7ANX+833V^W]\\6,>G7J>/PH-9PT4(2LS950@VJH5?0#%!(`))``Y) M/:LO4?$-AIUPMLS23W;$`00+N?)Z`]A].OM63XC\0JFC7%KY4MM?28C:&3&Y M489+`@D$$9'U-!I.K&*;OL*L*3LFDP#_=B'?\AGZFO3T18T6-%"HH M"JH[`=!7(>!-,\NWFU*1?FE_=1?[HZG\3_*MN7Q#9B[>TM$FOKA%+.ELH(4# MKEB0/RSSQ088=*,>>6\C6K`\7:E]@T-XT;$US^Z7U`_B/YAX)!%3^?'T%!5>JE3O'J6/" M]LUGI^H:\8M[6\3+`K#@MCD_0<#\36WX0UZ]U=KJ*]99#$%975`O7/!QQ6C) M-IWAK1[>VN7Q"!Y6-F[S#C+''OS^=1Z5?:-'8R3:;;2I;F7:QBMG8LV,]`"< M`?@,T$TX>S<8J6VZ+]]JUAIAC%[9:U?-X@\0$P[VB)$4.$+$(.^T#)[G&,UWJ:GI^G:2SJMQ';6B* MF)+=T)[#&X#)H+IU^>4NR-:I[6Y:VDR,E#]Y?7_Z]8>D^(+36II8[2*X_=*& M9I$`')X'4\]?RJ$^*]+^V_8T::6;S?*`CCR"V<<'ZT&JJP24KG9W*R(\6JV` MWW,*_<''GQ]3&??N/0_C736=W#?6D5U;OOAE0.I]C_6N3TJX*R&W8\'E?KWK M2T.3['JEWIO_`"RD'VNW'IDXD4?1L'_@=!UTY6?J=#1110=(4444`%%%%`') MZ?+_`*!>W^?FGN)I0?8,57]%%8U:5AG_`(1*(=PC!OJ'.?ZUDSSQ6T$D\SA( MHU+.Q[`4'!)^ZO0YO4;Z;PUK4UXT#3:;?%2^SK'(!CCZ@9P>OKQ5Q?%^A-&& M-Z5./NM"^1^0Q45IXITC6+DZ>T4N)OE`GC79)[=3^&?YTLG@S17DWB&5!_=6 M4X_7)H.1.;UI--%"\\;"9Q;:/9R37#G:C2+W]E')_'%1SVUYX=TB;4YM2=-2 MN&_>`1H^]C]U M-LV]KE%P>"W\1_I^%!G6YH1YIN[Z&"JR7$ZHH+RR-@#J68G_`!KT==!OK&QL MK73]2N8\2`3F2>M8'@C26N+\ZC*G[FWXCR/O2'_``'ZD5Z# M03A*/N\SZGG_`(YU`S7\.G1L2ENNY^^7(X_(8_,UK,W_``B7@]0`!>RGTZ2, M/_90/T]ZP+9(F\=RMJ,J1)'3^%7?&0NKRT@U`ATLQ)LBC9<' M!&?,;TW8P`>P'K02I/WZO7;T'>!=.\^YN-4FR[(=D9;DESRQ^N#^IK,U\MK/ MBZ2"V^8EU@0]N."?IG-;'A[4BOAG[#IJF34R9"0%.(_]LGITP!ZG%9G@^2SM MKZXO;N4"2)-L,75W9N/E7J3V_P"!4":3A"FNNK-OQ5>_V)HMKI=DQ3S%V;AU M"*.?Q)/\ZSM&F%EX?>#3(VNM5O0=PB&1"O(&X]%XR>>YYJ#QA9WWF6^I7:E3 M,"IC'(@`^ZF?7&2??-=)HFO::F@V:&X194C$9MUYD+#CA!R<^OO06O>K.[M9 M:&=:63^#_#UY>7#I]NG`2-5.0IYP,]SR2?I5/P+IGGWLNI2`E8?DCSW<]3^` M_G3?&8OI8[.ZNP8D=F6.W!SY0P/O'NQYSZ8Q6OI3%O#]OIND2!I6AWW%PG(B M+#)&?[YZ`=L9/04!%+VJC;2.WF<[XGOGUGQ`+:V^=(F\B$#^)B>3^)X^@%=- MK$L?AKPFMG`P$SKY*$<$L>7;^?YBN7\(&T@U=[F]E2(6\1**W4N3MP!U)Y/` MYJQXR2]>>UO+I&C24,L4)_Y9*,8!_P!HYR?R[4$1DU"57J_P+O@/3,M-J.IZ5QD-G$OBUK;4[E1%'<.999B%W@9//^]Q^=!4_=I* MG'KN=+!_Q2_@HS'Y;VYY'J'8?+_WRO/US69X'TU9+R;4YL"&V&U&8X&XCD_@ MO\Q5'Q3K;:OJ6U`RVT.5C5A@L>[$'UX_#\:ZO2-/\[2[:R5"NFH-TSL"INW/ M)`!Y\O/<_>``Z9R!"TZB4=H[&SI-W//(=0)*VS,#;QE<'8/XSWRWIZ`>IKJ+ MEO(U32;H=KDPM[K(I'_H06L$]*V]1SY6G#^(WUM_Z&"?TS0>E"ZB=71110=X M4444`%%%%`'*6<6P:GIS8!AN)-H]$D^=?_0B/PKC?%5O-<>&[R.)274*Q4=2 M%8$_RS^%=YK4?V#5;?4^D$RBUN3V4Y_=N?Q)7/\`M"LK5+:BK;$3%2?F;!XQ^.,UZ?6)<^%K"6[%Y;-- M8W(.?,MFVC/TZ?EBG2:+?7"[)]?O#'W$4:1,?^!`4''1IRI1Y;7*?BCQ+'ID M#VEK(#?.,$J?]2/4_P"UZ#\?KS>A>$KK4F6>[#V]IUR1AY/H#T'N?PS79Z?X M:TK37$D-MYDPY$LQWL#ZCL#[@5K4`Z#J2YJGW$5O;PVEO';V\8CBC&%4=A4M M%%!TE*ZCTN*=+J\CLUF_@EE5=_'H3S3C=V-W$R-)%-&W#*R[E/UR,&L;QA>+ M#906P`,DK[CQR%7_`!)'Y5LV%N+32XH7'W4R^?4\FG;0PC4O4E!="6V@M[:+ M9:PQ11YSMB4*"?PJO.=+T^<74R6L$\AP)-@$CGVP,FDLVV&5V8+%M!))P`:Y M/5K][3Q0FI*#+&N#%O!`*XP<<=,YY%"5Q5JRIQ3MU^[S.PM[ZPU-)8X9HK@# MB2,C/YJ:$BT_3MS10VML2.?+C5"?RK!TE8['6KQC%=7M_<`2O)`B")4;YA@E M\C/O@\=/77U!1>?7'4_AZUZ)G M)R>M065G!I]G':VR;8HQ@#N?4GU)J>@THTO9QMU'PQF:=(Q_$P'X5MNOVG7= M+M0,A'>Z<>@5=H_\>YU=P0+G$=N".D*YP? M^!$EOIB@ZH1O9=_T-^BBB@[`HHHH`****`,W6+S3(+4VVJ2QK%IV;G^(=#^?>K]M*=0U.[U3/[H$VUMZ;% M/S-_P)L_@HJ\=Y'\1%!S.\WS=#F9[>2WE,;#IT('!%18/H?RKI\$\C)'M3": M"/9G-X/H:,'T-="2>:8PWVH^)H;R;3+Y;&(JH!A);:.>GN<_A71RM>7_[N.W>WA/WY)L!F]@HY_.N MHP2#P:C.13N8QPW*V[[G,ZA9O]D6.)'=1U`&23ZUFZY87>LZU7-+^U-/-/=VTT;.2P4KT]!^5=)WIZ@GL>*0H4>6ROL9O\`[+M-W_7!G@-CH:`5)+8P,'T-6+2U:YEQ@A!]XUL@GUIX[4%*F8^H7UD]T MFDRW*V]MC-R^#]S_`)YKCN>_H,]S74:=J^F7S&WL;B-S&@/EJ",+TX&.E4@Q M'''Y88>ZT%)N#.((3_=)^\__`5!/X"M:N;5_P"T==N;SK!: M9M8/=O\`EHWY@+_P$T&=1Z674F2"&TLTMX_W<,480/\A^)J+P[:W,^F?:DN/*\W M*H-@.0.^>W.?RH.2;C.?LUT.3\=Z;8:=>VITR!;69$+2F$E<<_*?8]?TKN]# M-V=!L3?EC=&%3(7^]GW]\8S[UC6$%N=8C-W:K>K$DGZTU7-W>Z([A)3@FZ*XSVW8_E3W.>TVZ@V>0(Y,?+/&<.I]<]ZY[0M0+:E&DA^8 MR%6SZ]/YUVPI[$4W"M&[5SB-.\17VB:LVF:S*9H%?9YSG+)Z-GNO3KR/PQ70 MZU:$(LL#R(6)#;7/^-8?C73Q+!? M$-Y*AEV<].,]>WI2N=$Z47+WEHS'\$^+KN\OAI&J.9)6!\B9AAB M1R5;U.`2#UX.@+VKS?0O"NJ#Q5#J%S;F"&*4S,S$#)YX`SGJ?RKTE58X MPI_*FPP3J.G:INAPIX`(((!!Z@]Z8%//!XZU(H)[=>E([$,\.2FW%QI$C$M: M$&$G^*!L[?\`OG!7_@(]:W:Y;47-C-;ZN@/^BDK.!_%`WW_^^AG8?,?^`J0/JQI-3T;3=8 M$2ZC9QW(B),8?.%)QGH?:@YVW)N2^7]?UL>->)=<_P"$B\6N\,H%L'6WMW;@ M!`<;OH22WT/M7J&JZK8^'=,CLXID,ZQB."(,"P`&-Q]!WSW-.;P7X:/_`#!K M88]-P_K3X/"N@VD@D@TJV5ATW*6`_`DBG'JTY2E=7EUU,C3+JWL[`ZKJ M,JP6T2DQLYP7./X1W_#J33_"EU'K27.M321/F6&@Z5I,KRV%C%!(Z[69W3*[B6#*.N.*?I31PV;@NJJ'+98X&#BISHFF&>2H)R:V;VW^TVS(OWQ\R'WJK#X:T6TN([B#38(Y8SN1ESP?7K6E0R: M<)*-I^?XF-ICB:8L1AT4@UQ/B_2I]+UI]1A!^RW+[]X'"2=P?J>1]?:O1H[6 M*&XEG0$/+]X9X_"G.B21M&Z*Z,,,K#((]"*:=C.MAU5I\KW,[1]6@UBP2XA8 M&3'[V,'E&[\>GH:EEF#3I%'(-P8%\>GI50^&-&\WS4L5C?\`Z9R.H_('%7(+ M.WM%VP1*@_$G\S294/:6M.QP_B33+C2M4?4K96^RROO9E'^K?OGT!/(/OBNQ MTC5(M6LDGC(\P#$B#JI_PJV0#D$`@\$'O5$:#I?F^8EHL;_],F9!^0(%.]S* M%&5.;E#9]/\`(CU:V^W.L<>&9?EQZ$U;EABLM":`R*B1PE0S'&3C/\ZL06T- MLFV%`H_/]:2[L+2_1%NX%F5#N4-G@_A2-N3=KRN MTNK4E!NWQLO\##M_GM711:581@!+6-?IFK+6T,D(A>-6C'\)Z4"=)RARR*^D MZI!JEF'+(LP&)HB>5/LZQN4CO`-X81N4W`Y!7I6C_8>EE]YL8=V"N<= MB,&IHM)L(A^[M8U_.@M1J.U[:%]=LB\8=3Z<@UR]WI5I/:ZU:30*0A6=6"C> MJCYB%/;CC\:UM1TF.[L&MX5$;LZL'#E=I'\7'7&N:!SASZ-''(UQHGB#2--UB%+I4NE6PU$H"60@KY;$\C!8,/0J.V#6QX M=2+6+"]@U:!);])FCN%E7++Z8/4#.<8^OI6V^GVDUJEK)`KP(RNB$GY6!R"# MU!!J9[.VEG\]X5\[&WS%^5L>F1SBG^G;^NW0?'-:RD6WGQ2LR',9D# M,R@[6)'4C(()]:D\.2M%#/I,K%I+%@J,>KPGF,_@,J?=35>+2[".ZAN4M(4G MA0QQNJX*J220/S/YGUIFH/\`V?>6NKC[D)\FY]X6(Y_X"V&^FZD;IN/O/I^1 MTU%%%!TF(GABUCW^7>ZB@=V=@EVR@L3DG`[DT[_A'(/^@AJG_@:_^-%%!R\J M["?\(W;_`//_`*I_X&/_`(T?\(S;_P#/_J?_`(&/_C110'*NP?\`",VW_/\` M:G_X&/\`XTG_``C%M_S_`&I_^!C_`.-%%`H_^!;T?\(I9?\`/WJ/_@6]%%`PG_")V/_/UJ'_@4U'_ M``B5C_S\ZA_X%-113%R1[!_PB5C_`,_.H?\`@4U'_")6/_/SJ'_@4U%%(.2/ M8/\`A$['_GZU#_P*:E_X12R_Y^M1_P#`MZ**!\D>P?\`"*V?_/WJ/_@6]'_" M*V?_`#]ZC_X%O110'+'L+_PBUI_S^:E_X%O1_P`(O:?\_NI?^!;_`.-%%`PO_``B]K_S^ZE_X&/\`XT?\(Q;?\_VI_P#@8_\`C110'+'L+_PC-M_S_:G_ M`.!C_P"-'_"-6_\`S_ZG_P"!C_XT44PY5V#_`(1N#_G_`-4_\#'_`,:1_#%M M+&T