{_hϣ@dH@jX3.:mu!KL/T6,u{i3UA68߼v,ϱFbkM};w:5{z#~lU*vg*~VUsl[Vujnu-;hl[s6IkξLEs[L?;83ʶexCNP{;.5GZz- 𿴽uN~.+S-{lrݽ<:]'ҽ\?N-dRoL&-k%Nw sOGFlo#y|2:|aOQ-ld<Ŋh)G7[|y䍔)xck;xgyݛ7RcP7$ɿS~k䭞ZѰe?DRQ.>OOƒ$#m^lm^iA$羯w2 Z~0P/nYw(*W3l3g%ݳ-Lߎ 4_^rzB/\oo Igd0{!G?銢^ڧg+Z JOӤqi/:YN ^]Nn+­~|xuk_:_wl_^pcm?iZ'#o7,_]߻?$FF:p`O)37w]özA>9cI7c?W2HXl_GΥ<)XGr,KP;#*^|RY+@TnN9.l?rá<5>D)Osʡ߼iOwOs ߔ-iFWǧc#0}S_`ɽA<|,|͏ɄP ofiq/4᭾8%׏U:=+3}P-pj-m4oӷ s'N#)'z?rM3Z7 ދJ~q'gmQ|ar髜?se[]{JpQ{w|yo+kIEQ E.8P/{|Tvn'd𥞬x?G"r"fɵ/ z~)ɋ+-?R]W~Pm'D"N]:/5۹>OGc{qGað{苛󟰟gDkg{c[Xm"9>'u{{'guF֚]gؿL[cAU+*r<|g9ʩGJ2:uo 3~vwvU83P5\_ޫkG'k|q;Q {t|.`oϓG`dWWHgܶ}x./:{в/o_xxܡe_wvu/:]>po(ןsW{ 6`J'G~AbOQœ0v >W?tG{_Op;8Keݿq|'mHԓߋ'#UMOdA?\;WGsMVzO=ZgO^_WN tp:.mw] y[msL}{=׹'kOtKzTՉb:щ_F_ ըkWѺ8? n}?͘^Xۅ^z6)9tdXF?ipxعZaWOViLks߇๵ǛS\N&OEl\>e ~ziK؞Fl$ǧ~,GM83Et0{;->iムK [oÇw.ڻwvghk9o6?fܚ]?\u'0~Ӗ_OO asBWr&4j$8Γנ?O_o?t6&?]5Nju^*ןW{u<`9gG ezqZ48@ xpNt_m?{LݚGPߤKnP[W b?O:4󉿯\O?1L;9_x(;ǻȓo?zGrzv[KZ帆jX-1Cd>yɓ&?eoB'>Tly84N00n0B.LCgMh/,?WOɾ>[L[E#cxl|P7٠yJX98==_=xf$;Vԙ GĿuٓGP[N[_>ߴc򿉥׉}ipmqݩ~O'WJ8|z|Ӥ:8e3metfc{! 9m G=p۽odro7~N&`G??L<4v^cW Г]? ӳYOʕ{Fet:˂<󢳏 :i14>)޿ܕ]yLӃqb29Y6?l+?͵x9>W3Òv1>^ϬI+=̾\_Ǖg{z[phxz~/Ó7.GOSiʸ)(`M>nvxhQغ|z8tLÃxO?f+s!fO=ο^)ExIp6LԨ2/&GƸV籓&mdgf>xUqcQl9I9yē§>ǥaƍIX\щuzh vIlc˸br\*t7&zbxb\h?v/z^~)Ę{ܚ6<9ܶz2`x:-[5%d_|)?D3t>:Ҷb"fkyK[OQ+I2aY&/5~u%ic.~x?VP|3YWANVތ[16ͻ㸎ӛo闯>l-p|]uRɻ>l?m|yw%o\Y|Ƶoݼo OA_4_<4ۯO?iF3w?ɉt7_ˠ[2wi$~{W? $ZnwNvɩduj?yaTpyGQɶ"{,4qM3KNFNjv;G Z+gq`~}ef54_}%iPæ-=զ!}ί=T$]V/>Dzw5Mۇg̾wzq_v;-wm[a^bzcę~;;On-،k}*7otjrߓ uӓ9jKMOg *AF~dz縕d\oƒ&m(/2''u:{q|+ʤ汽rx@ihv5~_b.U~ٴ 8 &W˷y:wir?9t=45,<xh^lCm_;ǴcZ9nf2=J/0g龎W=:U~Jt[H$'i{|8øO>m7\OxB̰gmZg:xkҟ/O.ε=a=;3_Eml4n6j4h{`LՇ9qn:s L ^x}uzobƗ'ǃ/0?iO9mMN^/U4N&,"?_qbpѿ1Y|\Gwq⿅dw?R?X2<$x_(\<۽ȏGe.&5IIvi:)N 7́E_ 6/{npwpv7o62qIn?b 9~<.WgZX>3emq|Oneg{C;(Dmxv#yx|1J]]o%ro J|1\ϴ.>ϝo/;kGvB]*N^yCYe.'Me2wݩuKqs+q|<w˃^y^κ{ڷs<\i] q]\U6ڕA{|[hlOלT`zֵfPR睂{pN Þ}u+V* o/蓪#y*p/*Ƨ-R]QOgKZYVbEԗsk> ydl9vQl=LΕ:?定_>fݳQ?_쯽?-\M|~v)fޱWR)6.<;.ivjױZ"Q<Ȩgno6ow6n RUϲW[qwŋo/?,|vKljv{`][ji^Jcpd/γJOoۍ7۞جfWV}qh(oOvrԹݻ,UFp^i o_85 sԽQr7 ^dI,ծLk 󋻍o/S‹XP^X9;(ۧ}\+XM/r޶r]沅ԮsQ+A=uVxj%}'1iPǸ=U:ڠ2Jެ]߷]˶rq/2w_o/7rz)wgkʗ]+wÞnK7ǥ~AQbxQ:7ߊ_>jW_r| uvlnwGJmSmЬ`?9n;.f7ٗĶv}>흺6^)oX=.8T[+]֙}vֶ_t޻^Ώ_2-s evq[?`3'mΏܜ-&3#м(tFFa)JNXw.-SZj_gVu5;qoZq|nz{|>kۻ0ں;{QΝFe9/ϜW2k盩OwǾ2χ{ȸM~:# H2Ķ]KϾg~aAVՎO/ ~J\ovmsB<ߟxslSmxw{V_(M"Wv]Uɦ[W.sy}˫~joPfKZp$uJo[GG_\>e[m9>s~Taso;̦NRi'?=vt63J<7Zhb4RsgwݩP.j66/Jg[gS|{_.݃QNou(K_>oTp+k7wZK]SԚtӧұ޽8[>jUK`k?~xv>^*_ޜ)W•rzzsM4ʵ\ihvt?R;+ǟ^>m}LE\p=|xY~t}v_ToXt޼IlG )RacV|sv{{-9\Yi^ܛй\ᄐh[o/ًy\Y+]fNny˿7Ӿmڣ㏟snn ݹr߻LZƵѴ`QˌrVMFl?r}HY6\/vM{ł|hgӖ8es{ّׅ5t{v;r&];{~at۽}Vz;>n6M6:ko.sVvQ۾ΞR;+?;bRWtYwNjg/5J{֕SK%?RZiͩ Wg⦺g^m?MÖɥ>mׅAxƱ;nQdK/Υ]`P;s٭doA}q.s۝̛/c_,}=>5oհpZ*_Fh]KS]̼xq6:IZa\\A'[cq:~NѼ{ocGJu&υF~]VQJ(]5m4]nO??_:OsvM-t͖*lv/~ܦT1j~ّt}Tlגd,_},_r~~ 9wgo&_J_W4%є>*ͽVSlk~R<ԷGp=znޝ=}^Լ:pIƛGѩvpռN^lo,|VӲ6QMčఘ*wgOË~ŕwV8-$p !bRžA(]M]w>Ķt6JE`(ފ㳭 C&+Zl=+GoپH׏{c˵Vrg@((/;Ծ7ūAy~v\s뷛ʛg/AcΆqWP!^$LIfb8=ԧUvaͿAWǓB7.ǽ1=+`#&@KFAa: 13`0!Гf]O2O Iiø|}_ߍ &m:`\|r<.66uvǏgtm0 F0fDMO?dݏ{?X[\qLJϵ| j>?s?twp:~PAX,̤2<}tj(\0Ό뜉g scKM xiCם׉_hG||2 fVLnDM|Bs3LnY_$ZA2sy6;^A kS[ڇ1Ҥ&k3aXq;dcW< Í9]q_#}Q0$[zpUټa%wW3= 8o+#5?i"^xK}*hgrz'ؙD܇3aö8?8=9m;6H-}i[?|ubg_~~/u&N_?~!sm86\ȵg$ǡ)#q?y¿>3/+LNgT{XsϚMiKU{{4E/ǷO[o'W<\ ]|ɕkz*rg^n7  {쟁Iǭr:~P2>0xSvHw–Ĩ%3}xh2 L{~?Afʝ:?)s)?vZEM& O2Z6f{fMŹ Lh\<8֦Q>?$Hn[ӂ$5A<7$qMt6Wayf")Xgs'~Ny4r||taϋ2?GdO{;x8\V;ǯ艙O̩Oh[y-{r6w57clHΓQatz{x m&y8x|xʊSƧ|f|0]\gPʺ}6̇c!Ӎ֧ibSJzxy 3xAm7t/׉?-93f}s箙BCDϦ1YMwdÑ,J'vVS|l䟚+r_̠aRx?E?7;? ;otz?!M?ךj}a?$-d& Ǔ'X|_=`wU[9&AU]n9~799yC1hR Ň7n3oWk?(Xst yY{H=vw6<0v^Sǟf^RiǗOoMݨ/ͫ?Kon?GJ<n[_j!K2xzsstH?^%],$o//_/\WhO.7®IW{zL/u_G >(xGs?x+goU*?nF.?T"3x{Z?nzw[W7GzԿbO쬯 T!Q{zNCo jWhٖy_ڥzG1&wl'6lNѵf5kfQI郗8lYL.f'2"5b̖'%,R?yr,z*u\(ic4+jZU-u%q#t%eJ1ӳ\*H/^CJRZQu^^wfğg`% le0[zYXnD1n9G^"XM,г iFV7ҍ[.='cɪjܰN&gkNq-;>(A~30{SWt* *Rwbq, K' DT)UJX<reVh!Ѯ-?Ur ,FV1w-QvJ7{tiQxH\77W)#w/'ǟ>{q3 DsdXD'Z閗d--'C/Ņ@;Diąb*F@9'BV"L>J)Q$@X"i9?.! X +bx_/SsJVOS .zad3rb2jB,I&?V,H-XXhJi! L7-/VAj ,|t6/Kj^,Q!(WfZ7rrMO+:%2;$%33dsi[V6TJwZi~4@tUEWUH [6 =ZI3TQdEjrR)-"+ctei s2 (ɴѰLLrc7B4 )r7`X00ok-WUn d3r(,\ʊb!H1"a:7DhD -45;;Mxf"A~IT?8N1TvI.$l^s܎0{۔n T^K݇LrbqamA'B/ѰVdXVҽ" k_!ɒF?D ҖV-j!P @ 憾Eaj*^jiϩy̾ Xjš$Rm(H+SB@%݁Х|ƫ2bb[[qG%/0[ljZyѰG4(cZ (o@V3-J\Aْrn[lmŭT՞kHyFca4KG4'װJk7L/My:M`Eu5DOko 1 `%ZM;9'Aʋ¸D!Grjz'IU JY KH^ &|4\l5w5*A Aͷ+v4GdEjrRY㯉DclZ\FJ|UkSQ,(Ѷo~1Es)D[\ geh W _1I#@:GRzY b1vb@,eeRZ\짺kCJMu2"hFs!fqa~)/ ;YY2I zEꃚ\TAN ^ bWK{NMͫfYE1>֛ooۺJY +s!{Mp:vFךT9-.%J=[=6H G׺J(6?켷QjnM1 {!c>:DifK/Eω̯~d|_R7x_>S M!ڐSUjsڊzoiMלJ[EIkU@8eɜdŦ0f@1 nmXۡc٠ߵn;F=)TTqRhEpaʺa80%Śr@+_ s)q@F#bV6?j8+E%uDž|JApBW t麺MxKm`il-\Sŷta#c"}ى/@Ť>ֵF^-Ujh'#?nу 0$xaɩŘ` lHV3"dzҺ~9=UvaGp1.Dh_+if^zi8 'U$.ϋ Ah}A,0h@^OlIT*-x d QX5 @>-e{F(_uDЍd;ߛ| %|I%Ůtcuz Vt?p6\|l!k!vD#j紃G KWRR?yV3Ufs^VzztCcC@4 z A xɜG0 \kHAMXHr"ma,^ibw1rtT%*=bWK{NMͫfE/OiY&\k-1ITnGӊ\Υ:h@HI QĔP ldt0p.DeNI@+@L(20;ng=.7ӺٚS"5!6&?z0-`V[/ӏcL&.eM3yO²(̯ua=PLU ҆$':7jūRJX zhy.uݷzY覤 Hd l" Ridxs:cлrQ,"g}6'VzT*RDyaVzwg`$U^73Hfgh+F8(d+ى{N3W7Runu 1p:Sh5:iE{*DkBVr/9L/՘1+Eꃚ\TAʉBsV)5, _a.D5\]$=XWVt!kB,SA:cW겐(kWUbsi,f.^+9)檙'me&@$+,C@+Me$tGp W' =CE kHcvYqjs X ٱj6am4An.k^lϴZBbPHf*1NvF*LNJ(b {mĂ1 V$&ى'VzT*RLxfC偐x MIlM4ṕ3Xh B0H@jj;kj*jnb̦xns')%%IݶkY@ѭ"k< 2FZoT:m\A~R@oltSQ1z @PB63k4 pf9SJ bd&ulezY *"q/Xdc'xzo;# @\ V̸{͉qABv˟ߎ~m+30"猸:,ʥJ+W+ɇQ8$/ ܈XTnڍn NT2Xx Ck:aI&X ݆io2lj# LI8-grf=]'@vX #`%5YL.f&Zf ,1l{2KȪ+yQJH"  ]=||HlZIm1 "cZTnڍn ?[O ! !meӋVKH }cr_ č1p:ShWfe't~8vZ],eEmny fݒGQ<һ SjFCXFgXͤ3M8i@"Vc-uCKj|tTV+l/vN@"4X0ij^:ռ(ƇEmb 5G/_S2fYnT,TtaM֜0A"Y (if^zB.-.MeD!K&vdYq\{{NlL`U~[$?p*µCנߵn;F=)TTa-:vFךTՄ" $F cD'q]J ̞ݴJ}k9f&اE'u9nhtX `&띜Ō[l/H1 `ܸ-\!X Qԭ 0NjWo󽎕US"(@lw)*.4VC1 "N"hh˶RiGgF-$f>*eޙ}ĕbEV8 7ءk/ DhD p-7lQB4=!h&6C c!Iu\ױR4@JV3[mr@@ȍMdEjrR\^jə?]qB/κZZ5D$"iql^]f)w{9k'/IUxzMpzɎ dqawqWZLwNoB&V7<9-[lekril-AN Yr*B7guszsmX]]vfD!QtRSVFGdJ *!?%+UŸUj7 =;Mpܭp&G0wiJiMnT;%i[rMYZ ++2"h5'p?mz%LE, k3dt=2< |`UkWClGmk Imifw}wN=-rYo^H4i[3؟T!V F.MSWT6I,yX+K|8c󿖖ɉgG b!{(d3ryJa@gˇ]`oI`E;uF@ёh9\Ž)yZZHjNcaѽâLA p$ə@%d6U 2"5b~1"-ZjXC rjVmvf.Hnv#0Ɛ5Y$ybIֿm(XzYWG.#¤{%' ILAZhH/uI%)KA-朑=K$DA`XFU-Hȩ呈DvV{K'O~6yMɕŜyS+4Dr{@ @rbDly9p~jE!XgEm< ZXl?% [>*.]*+ V"XxceѮ'GHm8 h׷ҦYG:Զ6V2X ia,a?A]bqJhZ:޳*:AސąLE=^ V,-H*} BC04R(|%g jAka6Ur"k|@)NC \lVC`'OJNds @JLg+)@Z9jV)?[zsyf\(yb5k%^WxI3, "㡁 BZR72913eyC/y#B,1hwjjèTMAHAMX Pjiϩy̾ KQ#T@vZe!XR&s%ed <3v Y5!g&VMp=d(@$Aِ3!๫:Rv#OU`\[NڐAxm)iI7V:* fa(ϯQSfT;5DTj IV9) h?R2ٮb b!ѽm7WdB.v)Vv;ߦwW7.b]l$0\"c3=yMZO \k֩3֬rͬ&3. b!J\ߪ&띜Ō ˲X (`%wA~=U]+]н>wO_Rpjkp;;)dET$Q)<3GN+^UszSi굛v!6y ;BKoz|5YRr;r:3t!]oWr.(}m cRYQ(if^z,EAC}Bٖ@*`n#DBYΈB8mٕYM, L@APCk: cLBxVcD%, dr5-7_o f@irR`eZJ p zfrڍ']7亮[ѓߔjiP츝q lHqf_3ɲ)@$B4=St9xgg{!afYjzg?gI ȋѴ[b,YVUd_I>U`5%zZf$TQ$i:ibǽ&띜ŌU@來R:M4^Q!Sy&A@MH0cVtoQ=m9F|{< z6Ckؓ& DT9gŦҞSS󪡙}ANF)DQ\Mˍ׃]X`H9~YmiS+݊"@ BV>[үn#@ZyKvx@ lvEqZB %ЫC7VUkj7)"]XbJU|մ5@ŧՙDu]5A+K!$z7~Rn˥~Er:vFךT9fe'iE$K&X;BKp=lU(@UthAaQ>K*sBMp'yAXû3ndƶeSk=zZi8ZSrzɳVsKmh!6=Zsjj^54/(,NIŅI޴(tL= 5s5(6@kh&tU*{mol>JFq^Wř V$ _CI6ѽӚ 1ac'B;iT}d3 Bѳɐ#qfkBkat:H*䮜*iYϦŅGfXhᚩnf,"%j)jy riZZ$uIUk*~ +%@R+0="hP >As?@l)۸NoH tZ@$":J Ŏ`\=[$F@ \"mE·@Ȃl BVZjGV*fmϊ;k[-Gely#.QB6d*#Dubbd(JԩFE66(T䊥 ry>,e$JX@f'cXNC,p-TfCwԈK:<9 j*X@ئ9F&)EK/v{Լjhf߿1;LO%i@Ȯw`]6n8@qtqt]K\a5d_XEM;9'A)⫸e6R=9A_.ƴ<f~P"˩\%Ax]Qj [ѳC zө-4)m% Tt@qBSmgtYMYM՜(̴1#[B\Ք7rb<4X/K=AΦΕT/jFS~k;;qCk:~qp\@D(z$Yٔ29RD};h+n^C7x|yj/m 4ؐ`?ѷєv=wf҂.tV;{xJ֫Ta&:A!pkr^˖oLVT(WQWD'mS*]7nE`H XMFnR)˰UL"(j孥pZ]Bp{okTZ^\VRs_kKSQ+Wә+1]Wć?~9>֩3֬rͬ&H z \AFQ&W,Us~@lqlkǐpjR'wK)f d%eR=Qdaa̠OkhڵADiFYQd}^ tx/z'gv1#JS! h`ᢡYI3둲_/:vZ|o rIt)!{X@r5[fGsvUՊ Qt"a%'Q\o^_8mhc*Xipa"DS[ia{L%V3WgL<Qm"0nW O}ta;:DS Cf*@m ^y#ՓA[ʍz-[q3zۣ#K3x#=%+U XgձMC'N0n7wkD DOF3P%aq0^z<;X@T n{bD0= hJ-@tMo{R6ݹy!ae{rl 5\ƅ{{C zf DhAG!B@>`5Nѵf5kf5AV0(T䊥 rNZΡ`KƢ\d~/2 O 1^b!vǼF GN8@,̀)(UŌvEn;UFWHnO5ˊG2d?ӈ kzY r>/.t$ "h; \dG b[`5wjUoǛz<=^Zhftu/'j[r "ݴnXUpan @7yLm8?`;qAQf:%kb._өxIt zHl9~1bWK{NMͫf [bf5YL.fLE"TT]kVSfV̴* @k;:d],~hJi-"*[PjZ-ov&}+ ^eP圊zZ?'˅YmsUv{;ؐB4BdEa@IbaT93>Y(}x~k]{^-UrٳMtӓ.~Yl9^vnzv,ty/ݗъzGoyKJ( "V8jEMF'eWDWAΊV 0R<;JXI2mjHzRUDfa<3 X4DP bWK{NMͫftJ Y;0 {mUⱘiY'A*!B^Ř%@؄z}PANs:+?ią5+RR6 z7(ƇAo}z@EVBK+x'2WDWM;9'̳[@DZ bF5%`%Fj4:-: X"ayV5yW˖XC"YğB WFiR͊?@ȝ5ד: &HŘ& \JO("# A Eꃚ\qZ~#0Koia&% 3Kr0,3ZZ5ڸ0!8:zeY+b+cHXĠ4nmwaFF\ mmſu;sVYwkqU-e~'+ZL@1 hwiޙ2Nߑvђfex缋BKCs3F߇^-95568@%UKVnq>@h+jAFULV:]ܴdjeY9!9k1;cj-B9}kH(DaֈZLN/Da),5{mU,_Nu{ Tߎn-njX1P$JXY EݫPpE X)p߽7UF/g_~ݾn.nߒbň{M׽eӯTVB*|C;rJ2% PDG^^,i 4vDFzOxջV:77!py吷Wި+xtjF~AI00K/S^,dMtIqZ˒BBzFi ("ޝx3MI:0.([wIV o0୵O7wM]mBmLE{:vݔ߱=x``ۤo7ow>y`O60ųY vSgxU5Tѵx ~nJpm`+j:@n<$BD`, |;Nl+l4ZmEX.oǍڢ4tٷ[>ik6"J9Ƈ@n/d1 pESX7G6`WZl< BO(||sq/֫4.>>[.T*'r|0"RCOug5 @e:>)9瀼GwGol#=.l aP97~R?_tnZcR{Q0D@%rA6;a%WroBO[ vqKiӹH@Ftˋ(˗b}<\̒y|0Nd"-U,E籦B+YOJu ur6ZY%NbV/';A= 96upZy< Ep87Ȫs/z9o{O0O-fI#K'^g%WAՔ,PDAo&[<: o%MLkQoDl6Z@Kd$ `q4WV4Z011|RŮhܞrJNeBn~zYbQ"pە{bO 9G:/Ers;7NWW]w;ͮgAexe1Gٴ %HETuj:Kgӭx雲q|<7~R?_tFkz3 vpKv7id댢vW bhA*ك-jx7k7bC=28ot128#%rLz\ #+q%am4ڵDWOEV08*dkxHǤD|ݔni5yFPDEϟ?<~Wr'@ѽUxCtqGZaxpSJg޻ɃաK8 :x_BNuxZo KqxR,=H,x|7ݵ@Gr8R`r,@M6u3fjw֊Fog`ׯ8ʳrZٽw+N#>_aܟo@nu "a#f]b?WcA/]Pow%(Syz 8R<[fWjX]ue0Zi2N[IeF.= w??0: esOf^my~8[۝!T h7'i%mߟm:|VۢP-Gi]Y%r&,tzQ\f3 nkat h`, xsg2KNۏka9Z+m;jٯw.i O +| s1(w֋ pB˧Vw74Cy8|ݨ(= ZQ2Z6)]5*t|wArȗkk٧WZ- &xMn?u4௯W_K߿ZUه/?tP@anx7GeV AKsyW @] p(B_j&@EYehӟLwݶ 2șz7EAվn.JuvSƂܟC1|y~*Wujs(itJwaw7NWW]w;ͮg/\<{@nM)gP8Y:{dЛҨ. k=2zNc8]4Ywۅ@AQҿvh3ۋ\l}|?|oǸs$&d3^7h6vg;+- l#U櫫~|;׃VXvpsrvk_˥v* Fn|~ITjfqW8Ǜnr`WpX]kZ̳nh '^ 5,acղׂPza *v:9N3^6:gI+EkێgeW_66ۤ}\6wgnOj. w o31(P/H_cժ۾n.sT@>QPt\( _q*rIֺڈˤmGlz> :,]wWnIEUnѽT(tх{")U,8jKX7S_AЕQҿvͰA/R#Js mt6uP@f6B#7R5-^͢_u(z_n/d1 ȗ\gx\I~(שGlfN/ *rIֺڄ gץ/˟<ޖWYtL}Jd\}ٷWA;QP@iӹj_RY/-ӊ;IO@s12Ǜnr͠kA>f%`BޮKxgqlW./֭vneIRQҿvOnɕyvǂz_@] A74&zEB ;o%Զ^yaNV2a~x5놼]ʕm{X4qtwSy-Yju{\ _EZA_{!RN#ʿ-Gi]Iɠ O}_+k!w#=H Vt]U4QҿvrPH ZktZ 8?O?lhtܟѿP3}s:;iew7x .r6t}sm U->8)/>bh8`Oh?j,ӋIkxj.YppԨ/nEp_Pv< 7wșG1谞a{|i"YYAή+r߮kmz~R;Oj?$X_4z%7<:ӹGnK/_WqDd<ҏᤜ|5Z8-9(d;6u}H{7'*]5m㫯|_N*do\.Noz}w%~J ~UÒݼBG3CCSNL(Ysw@n$0oUM/͚Sș.ۃf=|Ql 8m6Z{mYق-@<ofO|Նf>{9n#tGn{;#v y%h:}q-jd&ڜLgOպkq8pz>xsI jҺ/gge;Ae:e4oWrv]LlS_ao9]갵w; C8[s [ *KBJ>-:?+ pOߚ oB(,]F՛$ a|׿~-&ap!Wlv?jDZֹ`8Y4tuFQr4U oNOUO"T1h<]=*ƣd?5cig PDw (T4[\yaƷ_?Kg_]˛ c(qFgӸ\Ynfj^.A+vܨ-I#OǛ}Y/dv=|VnB(g> e,brL kS1#fnl\Zя<=b{\/˫eQiGOʳ7nƫU*i}-vz}z/mϟom']NuH7 `@>JQc1^LZóUw1)1'pֺ˽(>YuQgkU8iӃ0d'jw/-7{r2* eMWt-O'xMn?uC:a)@=l q|ݼTr<{1,_d'&<iItzyaݍʐgZ v";XJ\Uǻbvq=}W?_}Ͽ拯rRٷzz:8(O'Gq;kE] CG% u$ \]}%]\ݞ-]Djh|^6M﷭`sz9Xfb~ЏvKa /[ E~((|;Nl@Thn  !O$ՆG2~Ȯ > 7"P:7kUW ,.|Y~W'rG|zU΢b] xK՛ S3Mh`\nc&:+Z+jmTٵ.Gf,@!iySy{[U {7>_pgFD7",Y2_xt7ҙ(K%5+4Tvh4}|*O'Gq;kEAw,Ž'AQ6 +v.4.W6񮵛^d[AFmHQ}:ۭzQzhI| A̒]eu闭vIpdTN!c(1u: 7m1oJ/S^-tijBޒG pc0x^ GO<[fW7vԐRI=cG |ivPD(xVƣd}ݮ]'E'hv#ɦcay+zu<ߺkq|!H\z. wNOxYhu2]i\ɪ GޣelMc*f:Z]7WW_K߿Z@!x\P_.I'%#m )DIbX^3R_TEH~ٵQ9ϲZ7\'VL,J*c?<)݈ +b"4TaX^Uf')'Gq;kE?v8y FO4tuFQr4ap^$brA:v画ePbyU]ԚhA*I_uY?,?dOit'o)=+vb?/:A_j$oޛIE$ wѫ|2FOS٧z& 2N( lXVp{S _7vLc( i}}vJ 0ooS 9{vU L فIA(d$O>b:\ùpv7id댢*hJ oץ/Y<[N,.`1ǃ,r]-T_}bױFR2oeul $'×j!gOTh՝W/hxTwGݨTj68_Ԗ{?ZH/[ J.h͈s"E8Rei}cW8@zϯEƻhvyp9\'~f0<)otxlz& ZժD_Re3wUmb5vv`-8`1_e< k]Br,zp8΃ۃd08lfݞ]#[!'%]pU?QP@oul=P@Y] {^ g7B) p0 b0х^␿4\ _o.eފ7w#cxt|% ģlS[`Ty2ExA*@]vr3a|ݯFojÛ Qurf5+ - 4'cMW|(d$vb?/:A{XdB _(_ kk!gNO 8ε|7ۏc?&ypi5fnl\ZQЬ1})!nE| ?̲4<9°|]2(RɅ, j#?զUmObO@4tuFQr40߼?!|ݻnv;j۵(GƗ$@ON!gz&|s,U SP@%ݒ(M"ZoZ~Hɪh%b'7x0~[*4z{|Qu@XCnJJ4˫$":ƣdNj; ˿/U՗//. _~hy? & B!oɨȸ.hZNxM0Kft*<;U9r P8:r9ӌ8R2WjN2mgt7>%(˗ ʲ8-:Φqy0yɱ@8[sq|M^?)oJ7J ȗ;pojo^(v]*Buzoj\VfGyV4g*NhI#K'^g%W?])b7IMi\d^^ 8;(`q4WV{9h0]K($}I]x̵s 4n!u _0[rW Óߔbgi2(/e@gxdxgq9]'+?Y,Q)ySPDɪ]4];#u oץ4\$g2!<Z?ߎ*vɩ24ZYUmة^lԾA%ƋUuQkv O!E,p$#I#K'^g%WAU E 8b G1|9xlx݌ٸڝUkd##ǵo۵~^# Ȟ0w% r,$ ǘ;2 8Re}][ި_K ٧.NS#vPDqOBP<1ÐA/]ЪuƒM-K{/}UtqMdWx4Qq4gj_u\ G"Q#]5vu(0%xI#K'^g%WA֕y. ^U՗//G1 Qp? Ghp 揄j(gNL"xzaU'/~7n <$ 3< %})ϸ)*W&>kt+*Ӈêrqg0;}W/mϟomQ:\nN:5;ț9 @&xMn?u4;S cKU\Mi$DڰSL٨}5ݍ֯ܺ'OW(@K/_Hx*?iQ|r2ƃlXEgӸDlno.Q= p$[r6Oy{Ɨ<.x,JJ5䯒ɾmMWIiHh0W0ٌ8Y+ ZOU/P|IF@~ު/7=z[1_b%u< P :U4Ѡ٘ pr4tuFQrݷ+Cޒ,~-sxVcAlfFMZ^޷5++F]Ov'(Y3PyӿynHy+Jri.Ƒx@;gS?)C@,O\l18RL VoeY?ھQ}ݮ]'E'-g+]e< QҿvZxx^eT[ݪqݔJ?Gj)^\ΓyY=N.n|!BSMl3M,y۵T_O1Roo/Is>Hs/sR*n&(ΉxTA̒\,.s}7__׫ydWŲ|6] .Q:WT:4L"-U4n.ٮ> ’2$d u @U&Y-z]}R>~˯ٗ|w~by?[@hG>ٌ8Y+ Zƛ#8C]'qlq6@񜖃}ݮ]'E'hv{ap/",?.NY9-< 䇫{ o6 Ogxqpd$QGtQr%TNlnNO;@*e]feo120 $L(om ߄8Jh5WCޮM__.o*x%(?a~q\*iO|{\I[QӍ <)ۥO3[no۵~y ca(qFgӸ\Ynfj^.ǹ(qh$>ofOlOZ"w|Rr9?|7dЉ@_PvEk38;w+8JWY 8tAޒQ98<۪7Iqz\ ׳OQҿvރ͵5pk!|3nJ/S^ͣY2(M\$qh$X1!Hvrru6-eoҿ>]ϢEP d7[/4ZP@?N۟JQͦҋv,WlfKȝy.ENjrZD|&gaip-CKƷCB f4.y=YI1[ qK>7~R?_tVڇ*6vzq7N;^xKB3u 7ސ٧x݌ٸڝm〢`ϧz& Z޻οޭ~T0{ܐ7ܐٰ%NOjipP@ׇE]_Y|dW, _1fO!g7[Nw&Ys8Oռ9>,}ePo4?junCJy2E:[-&ݓلÿ<ݽKǝph2JAh% P@A޴O'H8TbyU]Ԛ|Z~8.țë]Gzy|Ѓm;llkI| zj 9{x1u:?)9|д7n.Q{1ntgeqv֐ʫM|^Oi|K]5j͒V|~9-<Ŕ{H.Fpz1i V:{l7%C^i?B]WB_:L ^%G {r]%8MV7_Γ|OφvQ]t]+с|]Wv}u>n_nN_U|g[>E٭u/ٴo>$#KVf]OOa(8ε9٧o-a0^:0|Q4:awQ]4d3^7h6vg9nڐkMW[K֛5Ͳ+ OJeߐ7}h3ѐV-]l"OJec o.yw?Oյ颐=Jf{bx}^L ,6ڱ<@Ty 'xMn?uӟ#~]~VNVh8K"8hHi#) <{kY7 +{r)#~-lNx׫8y0ٌ8Y+ ZjhPDj R2/aw:-'Uן?7~&,tzQ\N-<{6{'>UWKe@ @܆٪52;șy?Y?XV.(uxQ9J(+<.]>BڠiaMkgt~R*(}(bFjˆ(M 4d$sTn3clfNr [jo\\?jN*W(.vo|ܨRU.Wl1Hd2ȗ GVqT_|ݾn.4{a|ry >pwܞfSGc @VV7x PHqcx~fYF`_ ٵrA}7T;)T}yBOO<]cY5ZUΦͨ[jU1m(;i|p,(x4Ei]Ihd8A^) 3e8@ k Qmv-zN TGq}Wo٦>ٷ6ݫ%p~,]?4x4MFU-@]kPLwKv7id댢*huao۵~υ xQҿv?|?x't|"8v_E$Y gɼNGMAnQ2|;Nlpr3Vmr!rv]ގE#iDxo6E}~GX=|!^BOjǺ<Un/d1 x.IպCSǗ.4i7j9ײa\/j罸q(# {5gPO+(/g_~ݾn.ZG7{kᦶ˸ :-O~~3hua04E#uQ: 𤔌 xdQkw6+@t1up1 J.hu[>PD7/-o?_\LYMY<[N,Qܕ{ݤN3mn' û|% %jfWˤ.Pp6eZۮfM}\/@Hץ1e}эow8rM2u"R5΢q,8b (f K9@!UtlI6> !|}CA\VxJ`xG (^ Q A[9؍%e=yk.rzP r7Ű ZPyo/"T _qcx~fYF`_$U4^-UUaXDFouOj&wV;/?n o?{npS1tUkǃ(-Fٛuˠϓ*, G[vVtQ :QK8@ԩ{uA~%waI)|YYrvhhʨq;gY<(QҿvA q1.a0lfݾڡCW]UNty>i]f":@>|zUBoH:~h S 3ŃI|ciP@iS!g?NJV\n |% yY4Nn;eT8鍓Eo뒬h[L 8,@nv{Q!u$ (7cƋE9lM˃h:EK7Mwwût:}q)%rPY f{zjͅ%>bP," s=8wΛ8sPЮUCW<=5trOEnIEɕ2x?T glkI|ѹs'voR/ xHE Ys$cj08ƇJFb%HV8 *q.]t}VH#^@=._\vp=vЉnMfQP@ײ hHh0aSf}9ZpY=M`GѮ r_:U=l!8ub<.giPݵ{UwFxR];,Wfz/Z礤o@`d]zV>(to>Yad@! ,RA@&{fwomƿ?F4(";'ȼ5ѯsxߵL1~&xMn?uþPOU @nJ?ό *zzLgZ_h 7!JTMVWɸuv>퓤~wRto{r޷MlMo(͒hZqTpB@8[z'/I'@A,i\En5t1kaT+a8+7Why< Ӻy*>@ U3zx@G7ԴGHd$r]d3^7h6vg(h׫ '`MW~|to&s#+W&;AQҿvA q1.a'j;\TW݋}^ F9Zj!/U՗//((| UF|\M р|I%Ei]zOު` !r0A^_hPHޫX,< IT,v4\Ⱦ(S-8ˁn]x*h-Fd]UK2Wpt pg<>(ϟ@|zQYZ 6ыx݌ٸڝlcuq[] W@p,Gty1O GPک)p4dP@2j~*zm5wf7OnIEUЪwÇITZ-h-hOdi0ZY.@qbp1\۫ECR!l~9쪼 4Zq4(MBD×(o z3|!Pqk ^qxq&AH{ʕAkvIf-+]o5E+h0/޾ĒBn›TOIs?bOiӃn^#} xQ9x8p9{yST&׻0x렧8v:恜>l[7z6ܔf,#=X]d5-xx @'C, _7OZx4#UillkI|ѹ]5B>Q\ Px&t=VoVqF2 .jykS?_V-|I2{M7.GI5~Q~ٷ/ٟ??|7斵Eq9(󯻡?mݤN3ۭvv]8HH9)U|^G| _ûn(r@YUyA/:fXzoÒ/|Ò%=wKFX(x]yՋ|6 Ö09AB _D7axR*Q98|kҐOz\= n|0]c(w&aSíM|SD`\mt6oqhnP _BPofO ۽͑S H7BP;~ÑJk 5T9.ޚ$@vSIZy4^ꋋYw637eyy6kGab=mN'_l! |E|eq pL&@8[9^\zݔF Kjۉvٺ?.AXU.|Y%$oi2EZ*e(ϵ?,I=tO}on;l !{!LQj qvk7nJ?5q8-x\WrИ 0"f.OĘ]Tˆ=ϩaۑVk * w @|:ٌ8Y+iEHJ}[7[X P@A _tQ- r9Nv,rݞ/Y{zqy޺2jy%ǐ^ۯP W_M~~3hjaS* $<6ϐy4xf]ONnu]^,r ,.GvI4>2YdJQ[4FT7vk^'ivՕy;x[{'Tܽ_k^"nNzmP즳nN_U|g[>ysڴOzMQH&@1*Gx\WWβ-NI%ku+jfrZ֧?-kux0kE9w՚7i{vcGt*3gy5/ww-ər oo~31 T}zЦ酮3̚~t tA5[@zQ |njԚ%rp5ZY-(]\ݞ-]Djh[{k\6M﷭`rv &G 'FwapPlnB oץU4xz&߼}NrMVu{۵:ߍr =sXD`\mt6x*l4Zm˅s|vܨ-I#OǛ}Y/ڮvs^NJ<n\2Ks|ݿ֫vSzW6Q?Lߖxa9o4ݱ|p PL~mr9odU{> ,m!gp%,N4l*@<7zrQҿv]p|*0vzQMY::(}d(R9gM܏\lhmIxx1 8)a]{N (h~ԶmqvYKdzjk6tSҤ(Um{1wS4$1KflWŸ9[eO8n/g_~݃&Z NE,/}kY{|9(-xx>yn=<)U.G`5l x\rX#2*NWhlkI| ڝfxuַ,"Ax\VM+W+u:ofObW _OJ 5@x)n/d1xܶSs]*ߔ~*J*)d5,acղTΫ޶UI(#_P8zHh{>R\9ky{\QrOHP_4gY+N|kRvb?/:A ݂)>N"y+Ax\ץr2WL _w0lfj P@_@<+Շkn- C w"r2/ۍpd$G8V4tuFQrF!|{ųevuO1O XtkφjiEq2h5:a?WPo4OU ~#EwۍFv/T OB%YW@σTv(_ kݭkb s1'[@䮒h5(S!Z={!x\f]ON6ox&xMn?uy}߼ޯln_|\\֙ch"z*# (àMY::(w }ϯ??bSw??X5܏{*@1'MjWzgo/'[wx?oC*٨fjw.BOgMvPlG:!˯_~ ɪNJd$Qgb8p ]":+g_ܶ=>J$ [=gܶvۙ@ qQ7Y"KN=*ŒJyq(p]g[0Z2Y w;sg4\?g22W;4ūq+^bAUiPMԧOOGJΗ\(﹡1$q<E]efw:* ,(`#ezDhE }:EjN@{l1|CJ;'I*ݕO'HnW*3@OH?m-'1յUΨ[ɹ]CHw  tò+,NSK|jNM 0HH_jm]w3u7UTDg]M aU\jhiGIZh!Bsa&?QqBeZv@ '#}-\ݠ' ۇZ_Ic4qXӨZqU <qQFA5'8K'2lr}[a@Zm J@K2<[^g0.d϶ϟz(eT7eNy{Y^B77yX[D M΢f~o3tۛȖIcE,z3;nhJ,д>mDW hNd%@ѩpA}: ۍ(Gȟ>1oҟ4_;uSc5L;Ÿ{2,FIf՛43ʦ@'Zhk9>fƸpךJ7 %~Lb ̗8V+oQ5lG.kzaf`LֱMaddgdgpTuH$#x,W0i hSnǾQݔQ^d\X!m8^ͣZt]js211W_3.3AjGcT@ $(@dQ:mNtnpf@M{[K(J0 NQ⢌Ƴ@ݽ"\AE :j R;\0Olg<ѝF^=;@/fi,yfҟ4 ѡK;ϙ IJȞJ:-8iY̫SNI2tanN @v,]3U@: uFhѯU:n]M-wi< ]<\@ Q-ga'(]fv.*(q)Kwh>VE:4k ' R25 MT겉XeQN <<7a"1}(^'fNer9ZI2=C@Nxɞvܴ~ĕ! NQ<'UDf*8S7Y%d%lk9>fƸpӵ*R+D K|0xFgc@m<|w. -dJh}Zgf]fnύt$ rfMșh5qRrF ZPhJF8*idW[ws? -TMW5V{Rh(i} %?Ǿѥ;*{5d/̈^o7FN~ .|& 9l=zd㴷MR]tiYvn's2W#\@ӵޓAL+ fIכفp K Z9ƀt̛%}*İ5qog"OjiNk,>G8}rCvu]LGLԄWo퇯>gp @LsC} %~Lb ڙ/gq\VޢHdC /)vd *-%ZѼN NKF@Mr 7n-֖$rKB˂rb۵;+y÷bBD ,GvQT7 M" ׀@dFh|Klѵƪz֙Ys ŚOhz`S/ i:)a"{B2i(%]of5@+%gh!4> Q$;f8vh.Ll-'p Sgݵ֩9d 8:mbM+ H$;4g)<6Iuq9ǝLrl'#L =>n>}m7;uF<ҺkC .R# ,m`B<H7i Kcb%$<0trd{}<4WU$@!,?ap0պwJC h@ pidak_3.3Ak$3EH㍳eXQ0Kk 96:=qMCh!Ʋl?=D0Qݔ9SL NZ^=˼b@קWצN0[薛^aiZ_IcU}H56$%dh(u?@; d{CC9tFn&|@.jy^'ᘦvB˂rb/=8&pw '[rrMOwo5u[{Si#Zzc4,{${I(g ONsQ- @6F'2<7ʫWjXy_uQx Ҙr[q63ƅ+,:i}UHYv }p.qAgpF) FO[¡7 x"\܎11=ӟ/Oha#Y<'~:-t÷;7  љcG Wb)1Zϣ^08mT '#}-\Ӻ_ݎw925&`{.C' @h2ĝ&u#6`del[='٩OZڧk@GQhv߮'۽ۯ}߿?|]>0bO*_z>HOh%wFŠ na/e+ E_3.3Aۻ"hl9Fv:MnG@^ضr&\wo.9\pi7yXP<)bdt0 4 N iΘk9}3vLu104_tCi@+2_65< kؓrylԵ,{Z%Naףը2 յnyѶ͘&H^IPm뙿Ie  6ltiL }AnBq0"pYIj0P8GfDK/Ʒkzڽk Mӈ8NQWfj4Yn{F戻4hgHv˚Z襚-ƊYfv {gVL8g*)6-t( .Mܻ5Lf ]ϻk|i@0HH_ x*2^[p{t^@6n# !8OalynZjdl] M94m ppii8dakk4M}xcۇI@QwݤzgQi.EL g{ߴv +@6&.3qTuU1 x(m"A* [6DLU$y Ʉۺ4WI7jaBXiD,}:,[$\؎rS/ i:)-t1mttE>8 !!O|y9,Wvʍސ<q\@;}]lg5z+[8 A,6FiL }1 `hb7p2i(%]of~?DR (Qݔ9KutҘhtAU]XSy>o3tۛW̬qnk[ڵW VNjk᪗^{x= ȌhpmGγ 3 e.|̾ڸqx t?^0¨T<-‰ZA^Q^?x X\Հ1dO@.RI`:tH N.x:C/MX7ʈhrq -3JgRccuMAB6$JWY,n\ ]ϻk]ti̺ܞFaz*OL+ fIכ+Ľ-5q{ Ib4dhXYIS8dmi@ ]߿^&`ڿ8ɩslK5j< &KNms:.u 4#$\tٴ8^t_JiL }D]63"GB>5P<3).J%:۲jB@ٻ.$yNJD}x&cGsjXyvQdD&Bڦ֙Ys 5qǴ$>3?Dj6 uwQEٍK!W0+jUXH-(G/&*o[TWy֛e18#J~-_,uZՕ8J1/Rm 6,`£mt÷;8w&Q8i^MM^iV2ۑ(zĽL@}q qo-JL0S30r˦02GdFh|K]+@8d?_ZFuS}2Ln<:]F/Q0ˠ.JQhW?1gLblN8BBE4hߚA{۷wjH\@l4V̒7& '#}-\bd8ӧ(< ̺ܞFtK_Zt{7&8\/ 818CF;E)E_ar Н^l@EF fcoJpY $;:z>{0D@넫^RV{l3̇"3?F[µzx\9\8@..`ZE}elC[<:%q;bu?kM>F@q~2Nik622߯nXjtqmInΗd2i(%]of5lMl-'p]G?)hO*AUr$:m<0<_h{}c\pv=8xIi@ 7-6e_hǺL@l/  dzڇ&iI櫾2ͫŇ5To 2._N&g]2_ꤥ1|]_+'!OXԻC94_gp H/aS <:ԨAlM_}*ww\+7,w8d^+]ޏx?޻h"-iھ͗> ۯa<%(KhM7|'<]yaB [hfuzaf`LֱMad1x@9_p22xa-YkS7+u@HYDךYG_y8pVgQ8t&VFm:@+߯ekq 7}ofQP` H9{pf`. tRw]#/N*_;z{+07%>. W专r[X@;uDvMV4->yr~^9u ju=7 MUB~IޝQ KBr˥gi׊rH93.ٴnL'3m4{*o;1jN+g;VB4 r0I1Mny8dW=e1r\=㚈rт,BZLiI_QXju=7ҍp&оrD_%QQ_ D@j. ,U Zh橗5 r}͌q ^bKɳ'U7D+ 0! ,Ǥ,Vp @>{qZyBSw*1L&˭c/qr ihϮ`Wc6O_h^E~rN/8?p4hEQԸKEslĮzӺ7EdMh̆<^:F@ .7$vi7d  @͌hp]W'dUd"\4BpgL3,fyꫫ]'h'Od}dKcb#8݄zKv 0F㴟Y_QoNE1̶,^,@! 2̺ܞFx]SVeXQ0Khi9FqQFAp=1L E.d;%UVUTߛ)Sb?kat]~|]Ҵz[S/WԷ?텥,AU_eTeF&](4*Edx:qeܚIgpZ!b]MtU57)؝؟GYPZ E)o1ˢo cuZ}&'@HxIXYiϮ oBFy/,X``D/@g_h^EY>ݎD9194aUbT<#ui1VTM](88leT7eN}NRЉ.&lqIqPTưʺ |-:iEajgյ =zc^]ٚSԺCh 9IKcf; ' \+ZO1.\u=ãkIU]BB<7ԙkg:YRAOFznj' ,O5 8 xPRyH4h{G< ^l@ufe8H7t5MqE<j)XrOٸkg"9xkӯSq3]Oy޽#* YL+ fIכفt]?tDel7 Uhtu^#]C4f3+0kpYmf }j׵ZMQ^ -DOo+~.Il7zC} U*N#"ә/gq\Vޢ lG.kzaf`LֱMad8) 24ʳ.ON sa4"'\;eZFuS3 qb" \xoo;A+'ksjNvmCE ڧS7h6zqfs3P/n[tQ l(i9%dBGl~\9\X\Q9h#AjUYh! XWI lG;bs@ eFh|ko3j@gEh8VAWjNjXqQ΂:- .JEQMDEhE턮xziqh9 N&{<>(11SH5 g@ e>[ RL0 O! awZUQ-:<-dzѢϴ6"MNGu&Q0WTV} T<-‰pF\}j]$_Ny鯽Ko߾7#\/ ϳWdSs W̬qnN/@CdѤKZכ2_t_ U@6iijCԝiˉ9'JH9^F!*M(rN0[薛5 R2Џ/> t4za<tvBD}k$ZO1.\~}VV#gr5x8ޒ_Fh,ȖIcE,z3;ؿ(\m00rB od^l7Ym{ǮgqTH֩&a[=  qޅD<\sqdjSEyTyZmY+5NuT2-ʴ(KjSy љ۩ m$!g!_^gFh|KxzO;Jǡa] jL < iKy ${Ay@Bd${Ǟ>6]ĉ"u^ftmaiD}@N:Y,]k>L+\'@+ wl$4&(>dkm=}J .cO344-ƊYfv z]tnfgcvW4 hU֙Ys jD k ?sO?hN _({*^^,MtIgV'{GQv?iQy4e]z=}&QQi[/qg!$\Ԃ.| + 8b^E.> CV<<x:i6th 8ϧM\i[cix+/8@GSKh?@W囿hNm ͟xǯKl4V̒7[&;Z*MgL7-la&mf W,72 „@t@)Y^͛ة4uPw&~_߻ m sٮܕAL]4 NQ"ӼSgaHƱYbpYd%D,[q63ƅ+lџx6 &j\aWughh貞6)fX^n@ 7ل(w(fO6=x0$݈nv$!64p prq \7ifFh|Kxьx?h~;W4xh.>&$gFZ&*(Zwƒ'(`#N26}dN OWjQ0zQqrbDMzjQݔ9q;E iQvn7yX 1@4Eyg!:siw뾦æoEt:B76I]7 +:9Ķkx {Gl3c\5B'=@::@ݽ͠ ە֙Ys 4{_TMSwe !uܴ~Fq0EaB@YDgnxa:c'>AvF]Y15# َ\@4P:moM bd '#}-<ĝk !8F}%Uqj,oD=}J .a=wNj#I0}wC}h|ɏ'겉J@ ( Q FǤ,Vp p|9jvgzaf`LֱMadZO1.\ᙖv}.#-,: q7W0iϕ_DFyJ[ubM6j\LG/ V2L.(B<?1!4&\NcQva%Fq9UЭtX|8j:&qcf? <ϡRFuSڽbH_@יv <*CvI:Βp̽Y2Z \;Mo3뭧߿o ʖIcE,z3;ib[.al~r-r)7!'xJdS~ŧfflcT%:(*n[nzvp>\ϓ,16IkNQh*3 Fjw ) @ 2{f 8 ȥ&mߏi7\'fѬfUhPy^gFh|Kx8id@11qE } '#}-<]IF/@sq!(4!$2 „@r J%Nf8n$ X'{u0 8ϳ ׻w방ZO1.\6egl;%,:͛( Ytu`i;+G-t4ֆp@$\^\ XK:) ;$@.1(ufDK/Ʒg8Jn>Qg 'L#p '#}-i#/LQݟ~&",tkk<2i(%]of3=MxΘg<)^zfIzr}IGKGup!$>uk9>fƸpg4qOI:8O捌piƖ=1FڛY4ֻK+˻QnpqeT7%ucp̥4q! .-nN ֳlmekx(4bi@vVjv#>BeZpNjmaڄŶ/'  h9x޾k&@ eFh|Kxd6hq2H֙-xZ8$#~<]ehϽn ]wfahG?WҘ@r*rWQ4,<**N4R /4 /⢜)aO?<@֙Ys i_}xhO/=0HH_ݽwꍷ)J˖IcE,z3;x2ԟ;p\h Gedtn,^d,=0_cN-V{`=;z;ڮoo;#3L< 8}yq@^s-^XոldSN> t{Vƛp TQe5b<OotKƉ=8W%ՀL/@'O4uNJx :~3J`\Bzk"pY("+UD.YAB}3@EaB Zgf]fnύt#\[wPTHvȕ4h"lr/=" zj$">?7"- ׽#-ٞ<l(ڙ-xZ8o~@pZ-tz:5 ۍZE&42G⛢ڇ ˖IcE,z3;hbk9>fƸp ڻnMr <|4醣ej"Ɏ̞l"tVjv#J/KelaVnBqQFA@25 8K'E|-tMI- '8"gw5AN {G }86On[SwyMĴ-h 'Dd%8\/eH_@.aq-Sa>vz6տeEBM^`Γ B8y^ufe8H73z^"@Ҍ-{b7xH M+7qC <Ǥ,Vp ̗8V+oQАlG}Vfj4Yn{F戓j@7}LTpRW0i p̮=(A^-)90HH_ K74K-|EJ4ق 6Za6aIrElY vS7yH _iYF$-sQ5t5mdVh,sSrC.">Ѥۙ: >^rj 8K_3.3AmNNn^@!H0.(`.88 *" FQ1t rSxli+:[&@xyNmbyMg<7,:Ͱֽ`93ݪufDK/Ʒ'f)6:7]63 R2»}%7ȥ6)^aaօZehp# \dLxϦkZN*Q F<.- jṢ`8=b/`VOן{=|hhWFuS|7C&@v6F5$ \BcOvW̬qngco~g4mow!d@fݫ7@1$ڨ3[,[z59QG,̓:RWEY/ 25a98I& id;\;ُ/  gt=Mw֥i:;a tô:I5Cb!Gvk9>fƸp?$}% >T>@kh :"^Es!$q ApQզG8N]%^dueS#:32s{nَvMs< }a4?p'[@.ZQ;< tCLC<R(jm?qk۫F4* NQ.7D,zV,|eT5EU=0{ M"]l"Xf^v6kYCo#p}1h9~YuUZO1.\a#RXpE>Weyb;΍(8ˀ\x]u7g'E (dW !yrsz8~SZ_  0=ף # Ng% [g5kw^=ugs#n+,V^.D fz#vLBr?7Pݾez *b[\JW޶LfOhkC- Ke=-p E)N0C+VvـdaiD}Rxsՙ·Uͨ)gznGVϚ__N_X'uhf᭪96n3՗t/p.zI3GlĽk!M}Tz?k֧~vBю8(E#EQI  B.,oӾrC .+Vil굿Zgf]fnύtC@8 Ѳ?^vM..>\gJcUPM پ*GK){ڇT|hN, zH;Nnl:I`Ab38|٫gZnӵ^N(-M\eof \\gXZ>*Q pNF9'&z٨kYJqϵYƙ7/H ְ4Og@>vڧO/ ʌhlOL׌֙·Uͨ)gDžg4gQߡ)LiVv4vr71FEB @W̬qng(Z/'N&7ƔE@o~ۯ{ׯC pk,C/Q~ gQao[&xr$,[q63ƅ+<&"9ISPF]p&qΌhGw5.v: R2²P G^WV벉>WvC~%ur3wfI+8 ;i0-3 ӳa#u2 kؓrylԵ,{Z%8en/̛qnEr ZcJϱ46|̷6$ӧ"p|I_@2̺ܞFxV@H!$LêvfTTp m!u`5UX@p\Vo{ˍcuda>8 B5(ԽvkV4t#:3רհ-Y-hln^E\ш),:͛syhקc(-cz2,ye<6!Cl4V̒7ap禮 +K/kݞs ?smD.>{jbڥvJ~]yFy- '#}-<ЮUfƸpɤx+:[&YaD FQ;iYuӼ 8ˀrܻ{*&Ҙr/`(ʯ78dn GբՁI_a=әQId;,| ώ;Z=];#i TgXr_H&Mvʽl^R paw~&F9ܻ4W̬qnv5E5ʖ[h^nTyB@2&\a 61g">d8zgZ([&}0HH_ յ_NiȶS228j.@۷ͷ?tr깆s2FHP-K-M^j=VbQ_55ҳ鹦iNyJM8*8:Rˠ^LKE|x(n.SW_'O+@!:9?4MMl-'\KZ:#+٬ѻFhrZFED ڇ0 {F}^hw\8sٙQ/(^|hr# '#}KWm ޤ VUM@Y(>A);Zh;7F8.4 ٽf (feBi#|@ eiƇ. ={<a8iO?h1#yK[}<"pz̲ɲ_6e/qE@l4V̒7a:"QB<(#ߟ/8opt-yXubkq=ӷ;5?ڸ뾓K{Zt>7!R\aP a?&e ӈX@ F,[$@  *1L&˭c/֙Ys sIliL qJs4oo$5oy6@׿>Io 6ާg=@+oշ8RM u^L7y1K&A\~(ja:y,fộXBc γSԪꔩ5BҴ:iڇedy3mƪ`W w5`_ s8 YM@SS٨6:\V_Wi#iV (ETuшC8y:ZVN ']dZ1JDԟ0M TOʌhևOaS<_ps^XՍSH~.H& !䊦UuhFgπds:2 gwspH|.uF<h6ϻkfL;EQbM'=VVo$ѽ3J@IBӴrC.4(v37+ܷxG@ l4V̒7y=ך^ OGTucZv'Nv$&Ed/tM  ;@ Z6ׇ0^ruA PPF(w:@Z\@_Ko\ :0>J<]+[;u"$>'x>횷@ƶ<"E+UږrE*!g]A N.܀|mm^Mƙ!GNG~ 8!!.6ANurv}=t6^ԚLVQݰ bywos㚰rjaś?͏_3hRu]VEw>m>1\[q63ƅ_S @ QY M>y:ZVN8gҘH_/&jՄ4!-Q83u&ڙNc{rLH\֓'Z_Ic#~mZ[{J* ٻwf7-e-SKpQݟh'kUs] nv$dAo\eNUN&Gp/bOC6*KmKuW }vUlbr=u(ev#%!gɌh=SgruMcv tz @dQI6.>Qi48w8*6t٣ae&M-CL}|KvP}5zok2{aU] @ܯ*}[YԄ\:@2w&<r+N'_w|paID:je-ҩdk&,r*׋̧.^acAZ8ih{CiLtYFp4q/[ׄUjz5`\- YW# >uk9>fƸpob@|k^3MMdFh|K WV4e7O{l4V̒7[rz4!wztp:CYXya7͌GD}H6.TTzFH^h+^"UIjJ;qz"/ŦG.k@ L# XQh[}NWV 6?}8~~_b ))synh?HF5? zGl3c\ulMtglvyE\ػwldq:X`@>&ﺆ^'vM@ rbOFMQײi;VfתZgf]fnύt#z{srftzI}eϴ?OpQyϻ0 <7zC} %~Lb g̗8V+oQhꎀ\oMv:838 tL @7uOαSeWԸEEzz(Oj|љDkS'?KӔ 3*p,UtJ7&@6W\891E vg8dDL+ fIכفp=,rگ (ұ9S3 YM] B*e *8q&k'}O#g)FNcnr f-6e ̺ܞF]/p UUl(iL}ĕK-hJ ̽ 2ܐ-tu&ҧ,Gɓ~_'Eꭼk_7~~g.ŪŤbڈ{@6VUTDԙD5Sr\s- Mq-t@*2.1YLF7 L'N3JlM].j%p¢6ݵ@63y>o3tۛ8T\zD+ F>@dˤ`tFpijM6h8V}P:̬b; BC2=Pde'ieiLquwW̬qn>\{Ozh%t_]3}ȁ(pk9>fƸpEp6_h0 vB x;&@;Z}5h5NSX!NX9uv8yxizw R04F"3?F[gFHru1r~0Wo퇯4g ! :LeXiD,}:,[.E9^|ma4"?7)TP@k}\)7A$?MHXYi0YM9HE(879սl4V̒7Z= <>(^}{k1"թnc}֌C@}-4mG`FwmY(B*(y2"=U*V8 'Ԭ{NӺ?aն׭a:[yLɼv,Ž]*@'O8d?&NvNw+3?F[`$ Y.pTCRӱ'OOs,"ZΝ8wgӪjoo^O;vtv\KN1y>M3@?h>a=~&0\_4 $HLEt;<0\Vpq''Pk{MW-׻TєZm6q^$˸Ynf%tIQg\J<=uQqtiH)1jr2T{$F3ICer B4 ?Ƨ&m-'Y6ӂ4=c3+ eZNs4-Ti m8G'vm[[uD.@11K?U;u:PgXvr] Sr ZhxYvYv^vjVk5|ujkX5euێ1odHv ɝdW 'ʌag`*QM߶s;,Vqt2'=+3zCjnj6e8['tnz>:[&@,WC/dh{,O; < !y^2G 8 *Zꮫ9wg#*7w״&YjLG_ULo*Z䮞ΣQ2*_trkU ^C -:ۯt1ߢW$ ԟdž ֡ZJ3Y ڣ@dcZhwjQk_´0 O! )Tà#~Lb ̗8V+oQppa;BβJL0S30r˦024egH6::'g>[7x(y_h^EN,(y)n4v6%/ ֻ' |to6=BM 1z0fS؞rtQ!zg}ƚ̓fds=h'Gd4265M@A}>*]rKcI>H/./ }h˪L O :DPD2" =k z쎬ˆrӦjAԾݾ" JoIfr) `p@ (Hm pfdXuПCs;#=!bp09G @ȇč? }9a7st pS:h*M @0j+h!ĺ$yzv\?I6Rx[gx(M%J\'j4`a=]Wnz+ |q{(rjqc#g:觟yFӈ5|.֙Q9qndv?7vZMGzb:'rbMF&rl5]&'e8epnOvC#53smޠGpcOY]kT]<MxM7zhR 2KKM_~˷PЬH@ls~m嶡i3,Vq8mkm g+⢌0`Id@+`Zh2:o QM%̊JHt^h=_,^ t Ampb=#XWϯc8~r Z벌JLS|EN0>ݿ NX7ZI9pfq!T_($ *1H@&ֶ3[>IJ xd) QQU9 =ݩ^nj¦hȚ.(73{kz(.,4s6~a[zYPky\|[SN"p{|(Y33Gv:M _6#d0>ѿ':N{  %Mb ϛY/W+wQ(8Ubh>Mm5ugQ㦻/Mٳ,~ASZq8>[t+@Wɯa4"'XJιIގyH7,isKGY?D0ڨ AV %bidΌωt#{(TFU]7ToB㕿xk㍿-r)+ "E|cȖ>[J`R?wx2MK!iyҒ:ܮuHVh8Ҩj~sb ,>tsUTƚOK"@AOFZ\U-/_6` vϧ`_=EHg tY< , {ִmmqȞkP,!)$YG*ċg_UufTexNڢ d ;8̈n8ln- -벌JHoOKFrauEVN OXv BZ:'hEȚ6` d1>9j@?Hy48Q$9@Ubh>Mm5ug|xc)@QU9`=6H@K<u{VN" BeHw=z+@;<E`;sUTƚ0 t{|rr6s53sm京V,%4%1ςm㩺Lm@K&yzZgr&.}o#2@KP?D {NFI7vHK6-i+gάx ƙtR?6ALUr"pڠ:8hZe>tA9 mM&Ly~6ԯ64q*8b@?LY/Tm|M@#k0d^̗ \ d,o!^X\) t2ijL-Z{N@=&(%;w}N-ǣdzjSz2т 62<hΟt|&]="0=15g&&@#+t9slnf^Wq @, @-ncK:v$ ܝl38e$% T$7iVNN蠻^OoB3J\g1ENdeT@O Z-6kڌY3+=ޥd'bs5m[l3}\8ie`גlY\+螪#6hǍ@'V{@^Uną{Rw@<ZjtaCE{)tPv;^j-+7LdT6>@ecZ(Oh|_3*3s Ȧi+ה^,(`#_g߿b'tcc_ye{vӏa8aToN(m^J4t|qpⅻyQ*D4(d8b^E.gMRQ0KTwf9 }cR98/ 2pn #5}4[jBl9ӣMw_z훲S@mM&LSSFU]|@3:_ZZOK(bd@!Ī: L1qH ԟ칞"u3/ƫ80Qo(jp$":nTU={N)@=yB @+4UՔ#c@;C2 }YLR]Z1T2 YӞ*袋ge"sl5]NXyG_g~ N._p{ =s@48_l <ˉ55jtL-fT':Yi&Zzp(tOIQCw2y>|Ǝo#kMk/b v?K|͗yboҔWg*Lmz5luR  zye{/~ogִmmqs}KI(<G(W^o_}ӷ?V@ZmhZ U\5{7Ea"ɔo,7{#}Z A@~0F-4atg:32s<'ҍ욪"qnziL{㷀ٻHaWi^GLDof2 E2ίkj^ ='#”G@K6@KtC'MhS(@Ubh>Mm5ug|zLUO0k( QpPbF45Uus#k0D. 88.#e>K;4m[7bZӜ6js{pS~7l4mM&L"?Qz8gÒޭxL7}oVAzԏsM{_Q`7[S5Z t?K>^ťٺqe;t|qLg%t!8Ubh>Mm5ug9usk2R|v@Cƹ:n@4p}ĵz*bM/v;[O#?mj3 :\ڋraka_y7h=4Us7]$k~g؎s=-Jtx]X"ą9@4 Cয়ufTexN3 E>, *%t@4@w'jMӅ=  H] v6<+ B6 {벌RBݎ/۴%'ntד'x}wV 'zxjnu6KlΓ']] 6JMcBZ?+)aꪢ(]*SL^vItk bEeDt<4FiuE7 '#mpрh Ov /P.PD>W6Ɵ_ɶrh6t4^åJH9>@y" 9,eGQpfidy0>ӏL}܏N^n@<' 0*$z)ʏ2/,ڹ=.`:ʍ'/(&,AU^^Fq&\ũe@t*~wo~܊⺮/| Rtχs\tWaF0t1Y##T2Ӟ2 ~ y/2rb4]pԏ2ۈE+Com⭧MՑ65Z\ :*7XћWX[[wj |p龘cX^Sݿ:j.|/ZziM0]KߨEHOվ?ѐ䫻ki_ᄆOeOzp?xvo&W{:G\%Ne$s5))mӆϥ8 pIh(6Gn|)! *L$mR`ϛY/W+wQ(Z%VӤVSzfYf՝Y\'"`_h^EN,pVw4A@v_|u+1/[7iXuFV8]]e?.07o i^M: iU(-*JhiL K8Xykڶ6fpw{Bh̥8.#βO4FQx|g2~u{DWmE~/~#=#BXa={s~EHxCgyiQ =߿￿9)&n@Z^)=3a@MU$^" :TY ڹU0@? L7}o}'yGg+%◻$4~)58sviQ@!˕VkZ~2^F⽫)ZgFUfAٟ@stHdumFOյt4 ~ X1dC3! ±puƉUgXmR3ŅWM@twy_-&>=Wյpf|H~蠻ߖQ*pIP{ëL g ODXuRms*@,MGAqݗ^nu\Kǃs1Hq0]F|@6G[k7DLt=KuG;%R sga4/BEZGQeLs?84{|ay3g}k4ܠQ~4_;"^ p;cZ`O  S4{ħpUM-%XTx/O͵t^q STF3{Ė, t@<ĝ ȇBAYf՝YTO^y]C_Me#u5m[l3}\8v\yg UmR`8h?ofq\EHw*1H@&ֶ3[>n6ucɯa4"'"XU 6z;`@< qZy4J=jrz@^UnETBZ=В|14F EQ^Dk+ Mka!$5ܟxf{CǬj6&﯒Sӎzelpd g"|@,Όωt#{[] 0HɰBi Fk"'1FyPd嚰bAOFZnEGaYr P=bՏXO-s@ʙxS? Uj/L? 3Bsx:\ڋraka_y7:A뮥M]$k空冦%3뗦"NW[.D;o m~tC(&fMAx@+.7γ]~@?uRymmf֬ԣžtjg0-K./R벌YB#ge64wS!=I~xVkSNvC `N,H HcbZOI?,/ :~lhvw] 1ښM>.!~_<3xɆYS(A@7Sņ "822g. wKb+5ImF,Qݙ]tXvyc<l3f -=6pEdeM_'!(Ubh>Mm5ug|!il>*(7 2X&F9];A꧶a{E= i;>'@/% ד'[-ג G@;[Ӷ6Dž#;(ŀ88.#Z9=ۚrC9ڃk $ xrtGsyɁZ@44ۀpbQW̨8H73 "~6݂C ;LgIGC3VM? r4I)$?7@ ._@n4IJ*p^闓(!a@⃫ZG×"NW["X!犑GIK5//i-g^QM6̚GlhiL Q\GqYhbmajqQl6}pՒAwf1kF`/|7K)\QUM! =%$  #Mb ϛY/W+wQ(ҝJ 0R#ЧI.̖:32s<'ҍ\>zZY:Z(:UZr <ԝ,I+R}tվ^f;V@cp׫ aD?ӘH|14Fr$DugV ki(Bʨ˜[TG1@4yb,EhʆriaНHG j_3*3s Țj?ik1=VITqOfQ$$! ˦]_y_eT}ͫՏ?9kk;?yn9Ί|@4yU~U2Jqݗ^~GUXϞ|5m[l3}\8?OP8m(@=yB .74j77q΋8Ӣ$=B3^+,d7q!+7=z 7~"u|LsƥHdݵ2@_8/|Yek9&EO=h'RWb P ֐1 g2_3*3s w$UENc4-2ʵt0LEa tQ]:B#[j9'$E9 S33:fTS}5cCA q߄E; \8̼R7ۆ`bt{<[O:}2)%]=AD;8 Y<hJwi!H6A3 ԟ53)-xemH!@;ڼߍʸI'lsۚM>.Y;U婺phJ]za(rR".rM^˜T@49].Jo^~{$DugV 8`>&3 /@}_QR=q>1 XZ(Oh|9ӣMw_z}~Kf ~^2Wo|Ao~myyXoa%6ά_h&rLZ1)@qK8$Hwʜd8݁7Yexb?RUz/x77D4lU~1"o&<Tޔg%!XGaZaFOfk[M]-AOFz!>hikڶ6fpy(7RUuHb_&4W˯oʖIYf՝Y~Gq0H=<[; MbAwԎ >}4YtoRM}=\ug,2=jt׾XT{&^̗ \ d,h@әdͩ @Zga6UQ'frUmi<2_butҌ- ׶ǷkG{hzD+2/hENJntU:N #쏄"T(@QU9<e:UGk}:& ta$cZ'=w+ճg+7= g="~r1Hq0]F༞tA97q PR83N1zD ԟRյ=UPbnp (Xq0]FiCpV3{*-H"u5,0U0lXY4@0J ]כBݑa-JC՟X6Mp8ZEi,=lT;bsX @@(Y@,|1l__V |^7\U@(2W@\k5f'E Y8Ie:-P|EX2=jt׾){4IJ$'6@LvJ@?+)8gIqUZCǝӅ%+=@gYbZ/ |14Fr$DugV ki(.z=O.jkqT/++ʉcQCw# /^ut3'Xر_dL~-vt@ 3a1 'k K<i/p^d*^{>ۆ5ΰXU!)s♙ ^>ڻMxM7zȄ^D [ 62;gv2K>|Ia5Օ랤0KC[/qgƸhvU~1"?!]}./YCq |_&ql$zV5#Bhޏ8ͣQ㦻/MtE@o?|o?}Bz/~CdI٘.=&&td0袇7tUIqtִmmqd_6o$W/(&ҼNekc]eŪ J[`䀿ZgFUfA{ϗ)m10Mu)ZN~@%: }ӓ4xXZFbe#tQOw32 „@V䇓 V$=$onPm9t:-#iYA7ZJ^o?#1K]dECGr7V pS4Ɖ1.%TVOWK>|/=ӣMw_zW'Q<~" cf՝Y~)X B<WiS1MErU?zB^-]^2Ҙ@ npi@"Ͷ(QBx 8pn2 &YvOp 1+hE*.sjE^b yiVmkm GTOU@}(VhVW4UOx4 Se%f~%pnzI32hPd4,Fdm4AV`QIzx 2Ջ(Gv GxtGYA :>,mL}mi(V@'8̓tC @+D+< Q檙e#>R=KB0„@VIY4"՟78^VP;iZaFOfk[M]-\Nk2_ߛ5W,:ލ}>0 @ q^Ι8,WDe  ~TL?4& =(<n 5 Ea;X < mCgXq t@*s♙ ^>̍S?wx2Mx iyҒ:ܮu?_ ^Då/vڹ&X~/~#=#>=Yf՝Yz:sdu=U Ӏh,@;jU9f?Rnz-uS k@Gz&\jӒ &kiĿ'st*} ]L- @ T`)K;vr7?|7mM&=?5VbzI0h算ƒ|pf)AbHaWi^G6qcfen4}zn:32s<'ҍzvٿAڐO@Ӕ߲g8@ODZeDt#uܪ,&(%;}s9_Ƹ_188BfrFKmqtJ%gTMS8Z vx[0Q㦻/{2S)Q\GqYhbm&pfl{iVvRܿ18^Y|/ؐW%t+@~rjNjWuR l,7_pb=#u \mCgXq y}<Wϯ z}A͍>2 „@VdƷ`)q2Kmkt\~]^܆AOFZ4Rϩ"E^yK pn7iVN'&fzԸK}svXXI@ǵ'ON.ԕߺ:32s<'~OVb@^e󺣇CW&Ia3 :wNs-B7cD{0>@7?OͰ Mch#F+ %! i &@WP퓑z"r8ZB3k돽.pVdXdX~*Qu# ,k+ Mkah.sx8IK95g&l8+J7`k g:ll>Ӎ Jz$wH61P̘/}MFzAlM&LGܠlITnz^\Q&DK>., }9/uPčM׮;MXo3=jt׾߂t$'N&ϟK{-Ƒ],r)+d\Ch*/h̉Ԭ$Z8?BSMD ԟzRpfahO]Oٟpf.&(ۆ.ݿ@{rt^8P ]A]s{MI7iIo4ͣRͧi5(*٣8[Όωt/x;+Su}2>r;g[ ϥ5N }9 ^ @Pq0]F7=@wpmMs+5rH"x-׈̱ko8\ &hKBZQb3h9. \㱱tx{ q4ʴ ȥ?<xD:32s<'~lȇW= yDʚ6` y_B]t! =I! .[8.n8 ?B :t5 pfeTenXBuA ;Uw We -cjKp@ Y37J~ &t|A)t]w֙Q9qnrNM#h%%$ S3=jt׾)(SvzR;[O#?mjb8c:\ڋraka_y73>KahGjrӣ]:vM*y|Qw@8N @1?U9oeWqE-(QALmkm Gg_ ԟn_cؖ5pj$#~^ZQlԁn?~ϧK|p5ݬ(*-mA#@[w=v5i2LQ=F~RBkR$W?MŪFf4f>#YwO+ 0! mR`8tOxZf1#X #5}4[jBl9kڌY3+؟quT1 }'@~w @,ڻO:}]gy& ^fzȮHBVzZsUnƒ|14F@22`QZVn8bWCHxܟxf{CǬj6&s,tRyy06(I5~_Ϥ>q~l ę (dq]5m[l3}\8MU1F(;ʑ\&zP=TiMb;( yQ_5K-?/M\]<& ]O?i*M׋GBZMyj q~XeW2nO7"LWt+@3W ZziL$9횾/}MtGVFj4iԅARf7@ezԸK}sl( x 8@s #-SG h*gnM,VQ&2螫W|^̗ \ d,o!65Яjʱt@4|j*~^4PqR2JtQNNhz< pfYf՝Y¥aR?ό2na gW& /ÇrӓffJ=Z D;Zs#AZuYFy_)џԷMGH+qǴ:b0xuaNxtsuV _jP :GnR|J SFYuޗ w)A>=22'=UYG@B$h8vW8\aZzw/ǯs:) |Q㦻/Tש"?Zz̓2RzZgi(M`bXI #5}4[jB"ijDU3'ޝ)Ab4]hxmxVm$*'( y;n,3]&pٳۚ"b*r?6ALU8Q*|Le, ԰Vr=<;!@|l񏚲??fe_((H]Cb-VFrVG5v?+`#_g߿vz~+|MB?Wr<"*I>0 iR'r hGa_- |qt:aL`w(>rWQ*-r9 R2ֲ9DSI$)xVFj4iԅִmmqȦQFD;x*5冨bQU9qt@3Қt @!8/Όωt#ۆ1@ K,Ua]cᾔTp^O:_/ɚ6` dG7v.m(}OSmM^=b%/:filugb6ؼq"tQ?LgIGC3VDR XG7ͦ"}O z0hfe_(Vnid4Ԇ$էDn#hG~8K}0;b}IKepf_@DKM3aHψ idtҁYZ>wbiڡ>-Xye%"vktt44hEDg@yfp FMOJcB%_r8ZB3kSC]3/5z0EPQ;FQUepp}$u{"U@,mR`8tOxZB!XwkZaFOfk[M]-gzԸK}S6MM>|:@n6e6zP p _nיִқ|+_Wx߲MhO~y@_~;|@;L7y=/{ql⥛2w+wd-ʑJ׻=.[1˨*E0`]l[5eUNjSZp++((Nh$gMRQ0KTwfYpҺ44Ͱykڶ6fpdM}$_C[/qgƸ46t;*xw?ikuZTnpv@1;@s tVFbgU3|sbo?|o?}#zAo^| yF{osQ]f-QT0RS]tr}x2c1 Z麱b48@ Όωt/Z n訔 t|q5m[l3}\8"1ĺ=F,4WM?_7q΋8ӢDk.WZ YjxnBV@4h=t|qfY|XQ?JO5MR!s )W0HɰBÁ 18V5UM ̇<(GrM{/:NچHeTeN|+FO>38\v|g]`A?i}Vf@mw@K:OoQut0P鸣V>rQzx$c xJ{s6@'ǰ`1,:_ *=nWaL)hd .@W<gu{\MeUT*:XIfcMYo qAG!=3BZ9=ak3B=U~l ę ($lA*|5h' ԰Vr=<;f <襵a+{\鏼5|,y u = XQ[rde4e|w_]K~o~m?WQBzaӟX6Mp8>髐@)7=)%`N t_N󊣒`y+Ms4!@%Q?Y%VӤVSzf,m6tчGYHeTeN`+8bs>WLe@H:USmC./?c}&&F9];,Gakڶ6fpd0wˬIj3 fά@4R~XeK@8bsP4A/n*. Us{xjU9FA%`_ʻ"Iۍpwz%E"G*WDx._J/h:\p:?ZC݊4'ó*F<#:4PqR2@g79voˑZf5/zB(&fMA8ފgm_WziL,Q\GqYhbmj&WY:/sJx{̉ VLM*tO$@+aP$Mb 8rr"@Ubh>Mm5ugûm]&Ng7v8o'l,o10՟tüόQe?פ'' :UZGϣW\Ịs&0Fβ䎯eGLzϟKy-T|4U:M((2Z&*{{RUn (#_3*3s Țf ›>GE.`8hZe>tA9ȀOVn8bW# rܟxf{CǬj6&+7[0Ƶ=T'n:)ŴP 땫^IIxR.|ε0I˯߼~L7}oʮǽpi)ne5ImF,QݙȚYMO0HɰB38yh6vr 3w&yzZ~(_5K-?/M\ Ҙbc@t:ٱ?@?JixOmqhw*1H@&ֶ3[><{:t@}d* yd:zUUt5eI?rؕl; ^_^KۮV[A? Z4M6hVe{I蠫WWC-Wnz'uo} [ pU% +WS0HH[V\  S[rJ\>Z86k=Ǥ"@$8ͣ5m[l3}\8RD@o?|o?}Bz/~Cd{:W5MZ(}E#Ʈm+a3HwTIgV'n"4d]ti(V@t@,YϛY/W+wQHS\S*ϔ@;M`8Z{?P8gW맟h wXOYD J:jVW:LL/xFӂ HށGaM/e+ :H>'ֶas}AO ԟiyf돦iݦ=@:8y x 8|5z8zRБ 4,`G<t}o0@? ֭ZgFUfAt$̳|U6(z5e H"XS@; xd&;zw2@ɷ5nkߔ5ݷV~vkڶ6fpdWA_#@,Y]T]wZ3|x꩹Y $!:n_r<冲 i,ǃmWUXCgAwT0m͍te _(gF=/ x:"=.MFZ>ê.fwid冸@E^yHvH`Zh@,:% '#m-kjOa'ۭid‘]ާj`܈Z=oɴk\WkͱkUQv2 `5nkߔ5*E$UЊ%*7gG+<'dPso}e1y>ы㫗o_螟|ǿW4+n~tדkhb*|,cc$dXE!!y&]=EÔtp {eVę;/\Ora'H{m$DugV {5ge%_p! =O! )ܽ tOTaB{IY4f] lX%EN\{`sϲ<N7MVFj4iԅAÿ4@,;oz=UGS ZѼJX@30Zwߓ:)@7-yxSo- Ua"+^oU%%d|a+ZgFUfA R ĿKHeTeν3"_8ͣ|Z. %@4uW͵jk؄g@ 5.@wWs?֕ hTO|Pbgqhˏ(0HH[ˎ+2) SInpC4 Z<E5羷'=aP$Mb /,(s#X #5}4[jBl9ӣMw_zH[>|r>[lijF*-r"Ö6tLtӆC6y@kYYpRrd){V\@xiIj3 fά`?Qoz~TǪT[-^Ol=*Ft?u q&* t@:X8hZe>tA9h3ƮMMƱ? ZH'g*hۃ2"?H6 R2{G?hKB%j>l@r,DoGxiSuMMdG` ^̗ \ d,o!:=yufTexNY< 9E֞k8Z-i!SH [ WD vU~1"N3M vy$ ~qVOԙc+*h珄"׋#7q΋8Ӣ$XLF[GylVb\č:r7^o駟q{r0:[]k|:/ &?6It*1H@&ֶ3g̵l)-eMRQ0KTwferCX{~Npʑ?0?7 qc_ho@ԯ64q*bq33:fTS}5Yٽ3gnX͉Aޗ1ſn8ۃ0M=4+ ~ uT@ZسV(kf(N硻uy'XaP a&e ӈX@,(t'C+ilm = 5Oc.t>鱛~7M~y9>t{S2pfdXEq@kMm\Upm-+,zRKE\F46b:r; ZUu3*7*ziX(I2-ʴ|:ٳ\j(A% pk{jO=,֦/4OW{BZIcbp^# E>((җ jSנ:bZȚ6` dʵtT@| '#mOZ|F}b"u #@^SCm2?keW. y~h1-1m6gVup -@K2 n෺ 8.#B s;o: 7pf=u-E9k9Znz=Gc7N9'WGIVזT]Zv\ 6y4qLLYznQ6Áf8?~_f;_L_n2axHoKM{z"VI:erth`YR:KI%Xb-C?G~8\ }D(*0HH[L\e?ؚM>.IshVܞ v{\k^F*6?FaQ/ Y= H4/J=V6³9H<Ԟ+{:YM#gzԸK}s_ҔkbZT6.Olj-0 3Bspݳ]E !nGH|puy:@4-rUr?2}]WI ;-kڌY3+-Q~]t`P3{8Yz*Iq:,ERR thЏN *5cpwߓ(E?iiCQM60ZiEZ'm&oM&L/z޿8wy(;ɐeѤ_^JiFI14eM$蠓&罛a/ϬY?֣Mw_zڬ:>I!׳a寚m,<;& /v9gA]@wG5|SCduOյ+7:lSCcp[D'=aP$Mb ԟ78^VP;iZaFOfk[M]G&Gg;h"5t"5sW0Wi :n o갤[ph677sU Qp:5TwYˌW *bc ݱ ʍt_qhq"_M? r_RU:֋/qӓrZv4 t*~wo֙Q9qnmSOEHbzz!\^ƒ|ɦHݥ~˰L՗/g밇P_Swpn\,mR`8 Sſ&ÙZSk|B>`B Zќxbqx?kXfcK{lSg.&.ur.`4Ks-'O}\_lM_O !W̨8H7?` }5 VAeTe~ӓje6ecWfvtb5 =<Ԟ+{:YM#/uat/(\^*7M6䋓Yqf^8q#\Rid}U?{&MO8 &gzԸK}S4{`/,El]Ea _ ])̕8Æ$j)@{#@9#c,oGvp,3J2=jt׾_;SO$O~acX8Ԯ].P@z < IA3-<`G*)j5A]Hjoh̎HqoGώ/ga '9'I֋ēMkp }4@ %u"hyx7-F@G6Ij3 fά`B2I蜣lSe+ >sǃ8LMDφC:ۍY,xN;MtC:d( R2@|>RO8a=Vnȸqq&g䏆(`Q0 J1yJ0 F9fy&E%F&ף` њK__!<@-?D'3(J!/3 $(_W[Z5z(j+L k Methods to tackle the pressing insurance coverage workforce hole with know-how | Insurance coverage Weblog – Insurance Info News
Home Insurance Methods to tackle the pressing insurance coverage workforce hole with know-how | Insurance coverage Weblog

Methods to tackle the pressing insurance coverage workforce hole with know-how | Insurance coverage Weblog

0
Methods to tackle the pressing insurance coverage workforce hole with know-how | Insurance coverage Weblog

[ad_1]

The insurance coverage business is experiencing a rising expertise scarcity. Whereas this problem has been anticipated, a lot of the dialogue on options is usually generalized to all the workforce. However not each job will likely be impacted in the identical method. As insurers develop, some features will want extra assist, whereas others will likely be higher primed to make use of cognitive know-how, like AI, RPA and extra. This implies some jobs will likely be changed by know-how, different jobs will likely be enhanced by know-how and different jobs would require extra people (an space the place folks can shift to, if their job is changed).

The very fact is that insurance coverage operations are altering, and individuals are the middle of that change. The query isn’t, “How will we tackle this workforce hole?” The query is, “How will claims, underwriting and gross sales be impacted by this workforce hole, and the way can we leverage know-how to handle each to enhance our operations holistically?” That’s what I’ll be exploring right here.

Urgency wanted to handle the rising workforce hole in insurance coverage

In June 2021, the US Chamber of Commerce launched the The America Works Report with alarming statistics:

  • Lower than 25% of the insurance coverage business is beneath 35 years previous.
  • Within the final 10 years, insurance coverage professionals aged 55 and older elevated by 74%.
  • The Bureau of Labor Statistics estimates that over the following 15 years, 50% of the present insurance coverage workforce will retire.
  • There will likely be greater than 400,000 open positions unfilled over the following decade.

These statistics paint a startling image—and one which requires an pressing response. However an ageing workforce isn’t the one concern:

  • Insurance coverage corporations are additionally making an attempt to develop, that means they both want a bigger workforce or the power to scale with the present dimension workforce.
  • Many instances, there’s a abilities mismatch the place the present insurance coverage workforce lack the abilities wanted to function in an automatic and information centric atmosphere.
  • Whereas insurance coverage corporations don’t all the time want lots of of elite tech engineers, they do want their justifiable share of foundational and complimentary technical specialists, particularly because the concentrate on AI/ML and the cloud continues to extend. This will create expertise competitors with huge tech corporations that supply greater salaries, extra perks and extra progressive work.

Tackling the workforce hole holistically

Realistically, the business will be unable to interchange 400,000 open positions one-to-one. And even when it did, the quantity of information loss with 50% of the workforce retiring is big. That is the place cognitive know-how is available in as a part of the answer.

It’s essential to emphasise that know-how is just half of the workforce hole resolution. Whereas extra administrative, redundant duties may be automated, different features may have extra folks (like sales-related areas, which I’ll discover intimately later).

Insurers must do two contradictory issues on the similar time: Have a look at their workforce individually and holistically. Resolution makers must know the influence of the workforce hole and the supporting applied sciences for every particular person job perform. However since jobs don’t function in silos (a minimum of, they shouldn’t), insurers additionally must have a holistic understanding of how adjustments will influence the way in which totally different features work together with and assist one another. In the end, there isn’t a one-size-fits-all resolution. However there are essential insights for all insurers to contemplate.

Cognitive know-how is altering the insurance coverage workforce

Cognitive know-how will influence totally different jobs in numerous methods. Some jobs will likely be changed by automation; others will likely be augmented by know-how; and different jobs might want to develop the human workforce in tandem with know-how.

Earlier than leaping into particular job features, it’s essential to know the forms of know-how which might be turning into increasingly more ubiquitous. The next desk highlights the know-how P&C insurers are specializing in in 2022.

Supply: Rising Know-how for P/C Insurers 2022: Synthetic Intelligence, Chatbots, Cloud Computing, Unstructured Information, and Extra, Aite-Novarica Group, 2021

Clearly, AI, information and RPA are main areas of focus. Chatbots are additionally getting used extra typically to enhance customer support, whereas cloud and information stay key areas for operational efficiencies and insights. Every of those applied sciences will influence jobs in numerous methods. Let’s discover.

The importance of partnerships

A fast word on the significance of partnerships: You’ll discover all through the examples beneath that nearly each considered one of them is completed by way of a partnership. With tech expertise turning into more durable to seek out, partnerships will likely be a key technique to bridge the expertise hole and implement advanced know-how at scale—and rapidly.

The future of claims: Replace and augment

To handle the workforce hole in claims, know-how will likely be used to each substitute and increase staff, although the size of this influence will likely be totally different between private and industrial strains.

Private:

Private claims is essentially the most susceptible to automation, particularly for easy claims. A small car parking zone automotive accident is an ideal instance of a simple kind of declare that AI can deal with—with human spot-checking, in fact.

Actual-life tech instance: Hippo not too long ago partnered with Claimatic and 5 Sigma to make use of automation to course of householders’ claims quicker and handle them end-to-end. From a buyer perspective, this provides a single level of contact, quicker response instances and simpler claims monitoring. From an operations perspective, this automation reduces back-end friction and ensures accuracy by figuring out the severity of a declare and flagging when a loss is recognized.

Worker influence: There’ll probably be an worker scale-down of the claims workforce as automation manages extra of the claims course of. On the similar time, remaining staff will likely be augmented with know-how to assist them to handle claims quicker and extra precisely. Trying on the Hippo instance, a part of its new automation know-how is to match claimants with adjusters—a sometimes guide, time-consuming course of. This augments the claims workforce in order that they will keep away from a majority of these administrative duties and concentrate on what issues: the shopper.

Business:

Like private strains, industrial claims departments will likely be each changed and augmented by cognitive know-how, however at a special fee. Business claims are sometimes extra advanced, so there will likely be extra augmentation versus substitute, in comparison with private strains.

Actual-life tech instance: Protecting insurance coverage partnered with Roots Automation to scale its trucking and industrial auto insurance coverage claims. In solely 4 months, Protecting launched two “digital co-workers” referred to as Roxy (for sending letters to claimants) and Rex (for indexing claims paperwork). Each bots had been capable of full 95% of duties with out human intervention.

Worker influence: Most claims staff working in industrial strains will likely be augmented by cognitive know-how. The Protecting insurance coverage instance reveals how bots may be leveraged to handle essentially the most time-consuming duties, like indexing paperwork. This frees up staff to concentrate on extra essential duties or deal with extra claims. That is particularly essential for the underserved small-to-medium enterprise (SME) market. By streamlining industrial claims as a lot as doable, the SME market might look extra engaging to insurers.

The way forward for underwriting: Increase

Underwriting encompasses each threat evaluation and product growth. This can proceed to be a key space for insurers to stay trendy and aggressive, so headcount will probably not be reduce. Nonetheless, individuals are retiring. Insurers should ask themselves: Can we substitute retiring employees or use know-how to scale up our present workforce? With the present expertise hole, that latter is extra real looking. This implies underwriting is shifting right into a world of semi-automation, each for private and industrial strains. And which means re/upskilling.

Actual-life tech instance (private): Product growth is a large a part of underwriting, and a whole lot of insurers are leveraging cognitive know-how to make the precise merchandise on the proper time. Arbol partnered with RealTimeRental to supply real-time parametric climate safety for trip leases utilizing AI, analytics and third-party information. AXA Life & Well being Reinsurance Options makes use of a white-labeled model of Verisk’s Well being Danger Ranking Software they’ve branded because the Clever Medical Acceptance Software (IMPACT) to automate elements of the medical insurance underwriting course of to allow higher protection for purchasers with pre-existing circumstances.

Actual-life tech instance (industrial): On the industrial facet, threat is the core theme for cognitive know-how. Allianz SE partnered with Cytora to faucet into AI-based threat processing for its industrial strains enterprise, permitting underwriters to concentrate on value-adding duties. One other instance is insurtech Neptune Flood, which developed an AI-based ranking and quoting platform for automated threat evaluation. With this know-how, Neptune noticed 400% development and is now the most important non-public flood MGU within the US.

Worker influence: Know-how is already altering underwriting, particularly from a product growth and threat evaluation standpoint. Reskilling the workforce will likely be crucial. Know-how, specifically the power to ingest third-party information leveraging the pressure of the cloud, could make product growth quick and nimble. Employees might want to really feel comfy trusting new information sources and AI to drive innovation. Taking a look at threat evaluation, a human perspective will all the time be essential. However underwriters may be knowledgeable and supported by AI and different cognitive know-how to enhance accuracy and make higher choices. Workers will should be reskilled to modernize their strategy and reap the benefits of the large-scale evaluation supplied by AI and different applied sciences.

The way forward for gross sales: Increase and develop

It’s not shocking that gross sales and its related features, like advertising, might want to scale with digital tech. Gross sales should get extra progressive as competitors grows and prospects demand a seamless expertise. New areas, corresponding to embedded insurance coverage, will leverage know-how and technique in a method the business has by no means completed earlier than. To assist this fast shift and development, gross sales features might want to broaden whereas additionally being augmented with know-how.

Actual-life tech instance (private): Direct Auto & Life Insurance coverage selected Advertising and marketing Evolution’s buyer journey monitoring resolution. This persona-based advertising measurement and optimization platform will present insights into the touchpoints prospects have interaction with alongside their path to buy. These insights will assist Direct Auto & Life Insurance coverage to higher perceive its prospects, ship a customized expertise and critically—easy methods to hyperlink conduct to gross sales.

Actual-life tech instance (industrial): Nationwide expanded its relationship with Amazon Net Companies to innovate and deploy progressive merchandise whereas in addition they streamlined inside operations. From a gross sales industrial perspective, this partnership helped Nationwide construct a Small Enterprise Advisory platform that makes use of machine studying to tailor personalised insurance coverage coverage suggestions to small enterprise prospects in minutes.

Worker influence: Gross sales, advertising and buyer engagement are crucial for development. Workers in these areas will likely be augmented with know-how, whereas groups broaden headcount. To stay aggressive, insurers might want to innovate and construct a enterprise growth ecosystem. Know-how by itself gained’t do that. Like underwriting, cognitive know-how will provide the instruments for inventive salespeople to innovate—and the shopper insights to make data-driven choices and promote development.

Roadmap to the longer term: A cross-functional perspective

As I discussed earlier than, job features don’t function in silos. So, this breakdown will get extra sophisticated after we have a look at how every perform interacts with one another. For instance: Claims and underwriting are intertwined. Modernizing claims to higher leverage the information utilized in underwriting and vice versa is extra essential than ever. Breaking down these silos will drive an enterprise stage change in behaviors and collaboration.

That’s why insurance coverage corporations must take a cross-functional perspective when figuring out how know-how will change their workforce. And this shouldn’t be a theoretical technique.

Methods to use tech to shut the insurance coverage workforce hole

Insurers ought to put collectively a concrete workforce roadmap. The roadmap must be modular, outlining which areas will want new hires versus reskilling. It ought to think about the interplay between features and the way altering one will influence the opposite. It must also point out the place folks may be moved round to capitalize in your present workforce and the information and expertise that they’ve.

One other key factor of evolving your workforce is early inclusion. Workers deserve transparency in terms of how their jobs will change. Early involvement will assist staff really feel like they’re part of that change—and decrease substitute fears. As a result of all of the roadmaps on this planet gained’t assist if staff really feel threatened and reject change. Insurance coverage corporations can keep away from this by being supportive, trustworthy and by listening.

Whereas a roadmap and transparency are essential from an worker perspective, the know-how facet is its personal area. This weblog seemed on the product and repair facet of the insurance coverage workforce, however implementing cognitive applied sciences requires a gifted, motivated IT staff. Insurers might want to marry a tech roadmap that aligns with its workforce imaginative and prescient utilizing agile methodologies to permit for flexibility and pivots, if wanted. Critically, executives want to have the ability to talk this holistic imaginative and prescient throughout the group—together with tech companions.

The insurance coverage business has a tricky highway forward in terms of expertise. Many years’ value of information is about to be misplaced to excessive retirements, and youthful generations aren’t banging down the door to work in insurance coverage. Carriers might want to get inventive utilizing a mixture of know-how and a reskilled human workforce to shut this hole and drive future development. The time for this transition is now, or else you threat falling behind. Simply keep in mind that staff are folks—deal with them with respect and compassion, and they’ll rise to your expectations. As we are saying at Accenture: Innovation occurs the place know-how meets human ingenuity. The insurance coverage business will want each to achieve the longer term.


Get the most recent insurance coverage business insights, information, and analysis delivered straight to your inbox.

Disclaimer: This content material is offered for common info functions and isn’t meant for use instead of session with our skilled advisors.
Disclaimer: This doc refers to marks owned by third events. All such third-party marks are the property of their respective house owners. No sponsorship, endorsement or approval of this content material by the house owners of such marks is meant, expressed or implied.

[ad_2]

LEAVE A REPLY

Please enter your comment!
Please enter your name here