ore) Sea Va the Pres SPY Sear La atl MAR et ata shh cau AT Ra Coe NA NE aK Sao sata hiv ath agaaecasy etna TOUR MESO E Nat SWANN. " si hhigea ty beara vilnes * Asgse AS a SP 8) ACOA Neate Ge ktayn ia FEA ot IAC WU uM A aN Date TA RRO Key TN ONS a gy tvs) Crk a ae narhyiiapectstence Au avekas DN OCS eve MNT Abaca Outen balt MONTE Te Aa tae wahey soa Vanity A ORALADART A. ry Pea tJ aiavas cAra et a raed a tba edie weak ou ayneae jab eathaah ehh A AlN. Vibra ase mh Syne at) Denn Garry Ve Ata eel rye ryi STENT EPR YF ATER OIDIOR Uc seks Was At as wae Nate oe dh aca hse RRQ Sow iA ROR yr) Repsartran Heo neat SA AAR AEC AiE gas hea a My mkt. eh apeaan Palate ata ctu sa Malan ate akyacbeve beans aN Tape baa tia rk “ptawmsasoa ace taens PMT RCs ORR CEE Tn TEC LTTE TW YPN Ser ADAG ORAS Sc habia: seas heati beer STRRIT IE MCT Mwah PaPOPE TaD ay Ua EEC AOrtTs VA Vag aha BSC to nny meh taeony Sa a Seana nce tame POM LLANE Dareicy ala 8 VS firhia ? aware! RE ATES Miele Aakokon Wgaphiedindy WUeygttin ts ake tea? ae SPR REDE Su TiN ane renin) Tyiget tener aah Vibes? rst Mite taantoen nana gh Minn Ta Aen as Ata ate tat mad CEMENT Rd hang “Ave Meath» Aca eR afta as 7 nay fata giants) a hey Pee PiRernaghuaar ences sctemeaneciioninat terns i Pers Satta a in sa canebein SD aN AS ae ANAT Cie Pde aetna ts i TauNaNuncragecadetaeanak keh eae 7S) ih Mack asa takers Tart ke Thiemomeey Tan nents oes tees MS eet sSiaadad nash ate ta Aedes acts hace sate Nema on ts seiploee! AEN RA Qe ER Aayane erent . SApA A Naa a ch Faved Li Noa Na wany Vi Sate ia Atenas TRE AMAL) nde anaey: AS Anon ia i ate By tetas ae Tht 4 wy \okaaeae oe yee 2 AY or oe PR Monat a8 08 tine PSA Sef sie beach ABA AS et RekeNS PERate ke arate cadens sich. Sab ancl aba chon se NiGTR seteauned shearersupe ath adn th SAE HNO a Sade lay asi Mp auth hn Re ON RT LETT Jaatgaasn ots ake eatin Beane ed a cien beam rectenaces Peceryur aren rear vt ¢ i Stated HERES scar eorer rere ORAL ssieceeatregiog " Brae aah poPinwoe bab EU 8 Bala Wf dedac 82824) MINA habesae hone, vats abr eh Paden Serna Bae i . SPRY RATE Poe ren prota Nails wat Ue iean pantera? aah ‘f pris Sa ec aH see de tan eae DORR, nealacpre ce: AOE ey opea Dy raat orn ou palate iors nied Nat) terettiaa arte ov Bitisies rte Sah Orarstee Ete yy 7 satan me EAE i aihaes eRe «Si ub Ky ihe gesattsaen fo igeael tage Magis revert hgh @ Wahi wie ween Bier er ice fo 4cie iasetninciay ALOT TORR Phelenpr Gaeta ieteyivergey Bears 2 epee Oauie Meraaeh het ts tent te Toate rela NAME IR Fates fea treguaad een besad te tee Bar Regan Tyre tery gu rigis eet senior yer Ao est nte Sy rctiettote co ieeetgier= i pusenecerirtpeginies cee oe Bite Chie seu teigieasirne dint Ceart AN ee yey er Hatibashs May Msakree Trae ost ey bopae iva Gat as Meus Tins cela tiig ta rkpea wets Pr ea eh gd art BL TENa INS Rory, TORE er a Sa : RAH uate escent aa tarD epi eneet lteter teey TAN Cewyor amutaenyeett (ited tegen iGiecelestceentetet sii ti the Pxrearres Tren ee Paine skyatronyeeiycasee Feu lyre! abee ste ae Jit wr rrigcone Froweevcc ty Shaka aerate Seyineeye alg 4 y be dey “ ey atin Mice eat f Sth ete Yuna teste Taye ciniocatye a VP ary | Bit, Ban voy , jv Nowe ky Desai aeme ry rrineret Ty 5 otatatitinnayh Lobe Mrrevecprelatbetecnealatae Liyey ese tS ete oped GAN ete ayaney shies Metin aan Ne Hyhaeiceregy ‘ia vive Teen ney 15585 POR ata! Piers ie ¢ AUT RS a yoy Cater ets tts eet wees ean % palal graphy wage, a Bante rergene ie a Fone nays - ih Teritacat iy eyetty ” Vavud eres avaeserg OT ales \s ¢ ras: WotR FEL, ByuryMT Pee aI. spose v4 8 Tremere veh Peers et Wo psgeben Ie (states ett rat ae vibyh yet rine n Aa ete cere freely oe WN Iaachobalyress (iat eh ay ake eeatad edly a Ra 9G acgeahy ey UNIVERSITY OF ILLINOIS LIBRARY AT URGANA-CHAMPAIGN BOOKSTACKS Digitized by the Internet Archive in 2011 with funding from University of Illinois Urbana-Champaign htto://www.archive.org/details/influenceofpopul33simo use er, Faculty Working Papers College of Commerce and Business Administration University of Illinois at Urbana-Champaign No. 33 FACULTY WORKING PAPERS College of Commerce and Business Administration University of Illinois at Urbana-Champaign November 19, 1971 The Influence of Population Growth on Per-Worker Income in Developed Economies Julian L. Simon University of Illinois THE INFLUENCE OF POPULATION GROWTH ON PER-WORKER INCOME IN DEVELOPED ECONOMIES Julian L. Simon* INTRODUCTION There are two contradictory points of view ae the effect of population growth upon per-worker income in rich countries. Classical analysis using the standard economic concepts of saving, demand, and the production function suggesta that an increment of population reduces per-worker income once the nation or combination of nations is large enough so there are no important economies of scale, and given full Gen loveant and high-plateau life expectancies (see, for example, Malthus, 1803; UN, 1953, pp. 27-32, 36-39, and references cited therein; Meade, 1955; Phelps, 1968). But historical examples -- the concurrent explosion in Europe of both population and economic development from. 1650 onwards, and the failure of France to excel economically despite its low birth rate in the past 100 years -- and also contemporary international cross-sections, suggest that at least in more-developed countries (MDC's), population growth does not hinder and perhaps helps economic *I am grateful to Gyani Singh and Israel Luski for programming of the model. Mr. Luski also gave me valuable advice in the manipulation of the model. I also appreciate’ useful comments on an earlier draft by Professors Simon Kuznets and Nathaniel Leff. And I benefited greatly .from the comments of the faculty seminar at Tel Aviv University. -eaniHo0a asiosavaa a necour sa - | | nome ah atta an nae ob | Aan) iF i ip a _ eaatitn eda suods waty to a2atog ‘erossiaranoo, out oma faakeastd .aatstnuos oly nk omooak roslrowreg oq dawora 7 F bas ,baemeb vantvse-3a ‘aqeon0a ateoaose basbanse ate gale AMetlgnoe sooube nobialuqog re eomtoct ne ids ‘eseoggua sotaont “nok ogyat et anokisa to noksantémos 10" noktan out Ae : a2 ; ania Lad agian me woe) solonedoogaa ‘nit's waezatanisit pis 2 ‘cobzolf iutoxeds bests eoameroiex ‘bus REE SET8 gn seer « a fokeoigxs saeytwaM0s oda + ~ wsignaxe keakroteta au “8aet age wasn Ea 20% hod 03 movong oun e2tumes on fl ota! aad hodiew wotatuats ants abacd roda0 943 sda 0 -auoksqmusanat | ‘erent bier roddaks oun a duoda dxtrosts ‘8 ante gatwolte +9 ee om coir 10m sab bes swolte aborsen leotoytade aed? febom 2 = ets acme ahem 4 di Aiut~tneo ek toqeq ait to axisn09. te : ‘viderovas eto elevates +e haat bit Steseyolgas ’ ‘ sroddse Stidig odd patbutaak (brace Bo dake ovisteng aes Re. | weganda wotue oLétauog anis oe ‘yeode eidaatottivn sh nosed ns 0% eqaiiteg otal - babrngernth od em aie Paty sonmoeen tinting ‘ve ' } ¥ ies vi e nk ulti testbed pauls nove 70 hates on to, ath aqne sais ih “= > Ser By pri pei er a -3a- or 150 years. Though the terms of reference are to the United States, it would be most appropriate to conduct this analysis for the developed world as a whole, because of the scientific and technological inter- dependence among the MDC's. The aim here is not to determine whether it is worthwhile for a society to have more or fewer children. Such a judgment would require various difficult assumptions such as the following: (a) A rate at which future consumption utility is to be discounted in the present must be chosen if a welfare decision is to be made. A very high discount rate implies that having no children is best, because the children will become producers only long after they are consumers; and a zero discount rate makes any meaningful calculation impossible. (b) A welfare decision requires that one decide which people are to be considered the members of the comity whose welfare is to be maximized. Yor example, should one maximize the welfare of only the people alive today, or should one also include future peoples' welfare in the objective function? (c) One would need to make difficult decisions about what is to be considered consumption. For example, which parts of a child's education in the home, BOIBIE beste as, cl ot sonsgater cal wines oe sooth nay, Pe beqolevah ons, 093 ‘shexteae ones, pauboc ssetnaorace ze0m oat fit woaat Laotaetonioas isis 2tzimaaton add, be nvened, ieloihe a et any 89M, ods. aso | 703 oLbints 20% at. Py ) xertaede onleroiab 93 io at xed ae od? -ouduper. hivow anamgbut 8, Hove snoxbitd> ovo, x0 stom, ove he By Hoddw Tt stat A (a) Latino to ad? ‘en ous ‘packs qmvaee seo ad Jaume Inseexq ats at hosuitosath” af. os , eatthaa sotagmenca 9 ee eiay jnuoseth deh soy A, eben 9d © ob tindatoob sxaiiaw a 2 ated Hiv soxblid ws eavezed 128d ot nexbLia an antvad ints 8 aie ‘ea Snvooath ras a bos jpsemuande o%e vera 108% grat vino pobekved exgiiow A (df). eidbanoged antieluotes SeRartanen ® * wredmam, ods bexebtencs ed 03 338 siqoaq dotdw sbbaat 0 ands Biase catquane 20% das hutinn ag om @k exni low daodu YWioumses s ete biwode 10 eyaboa avite along ata eine. io red Lone ania ec: 400 (9) froks anu? eviasatda od at oxmitow: "aalqoes “ae ales hs ‘betabibans ad ot ot Jody ‘gusda pavhexaae awoAVAah ovina oa bose ood edd ak motanoube, «’ "Lbs & do ‘asthe donde atone Bias 100 ry ae oe 7 aac ] 1" ivr ow, sh ; ye al Le ie _ } i op ake or ; r ed oe oe ne fy ha ; i | as street, and school are to be considered investment, and which are to be considered consumption by the child, his parents and society? The purposes of this paper rather, are these: (1) to understand “in the history of industrial nations the influences of population growth on income through changes in capital and knowledge; (2) to consider what ‘the future course of output per worker might be with higher or lower birth rates. ‘ The dependent variable is’ output (or income) per worker,* and Slot ‘consumption per capita (or per consumer-equivalent). In the long run ‘the two measures are much the same. In the short run an increase in ‘population through an increase in fertility necessarily implies a drop Sa consumption per capita even if output per worker remains the same, because the total number of workers remains the same while the number of people increases. In the household, income is then spread among more people. And when population grows faster there is greater public consumption of education and other child-raising services,** which implies larger taxes and less resources available for private consumption and saving. But in the long run ,measures of consumption per capita and output per worker will give much the same result, and the focus here is *It is not assumed here that per-capita income is the appropriate measure of welfare; ‘elsewhere I argue that it is not (Simon, 1970). But per-capita income is one of the arguments in almost everyone's welfare function. *kThe investment aspect of education will be treated later. *y sietrqozqqs sd? at emorak a3 basse ynids axed bemose aie ke ar ad 0} axa Waki bra snamdabved habroalnly oe tn ise oe ome Tyietaee baa astioseg elt bitty ad3 yt auhiquveaos bexab . ' baagexebau o¥ (£1) sonoty poe ‘ radset edna eine: de eitonpiie | at” ddworg nottatuqog to waoasuliind add. aot sen: atyteabak ‘to saad sadw asbisago 03 (S) depboswoud bas Taskgeo, th nogauds ‘dguot fowel yo sedgkd dttw ad adgtm re = uqaue, te intel bas *,xeskz0% 300 (omooal 10) sugavo ‘at oldetaay jet hy, tux gaol a3 at » Gaafeviupe—teaumen ts 70) aaiqaa t6q 0 jenn bs: "at dateiiut Me srt BxAne ony nt .9ane ‘oily Howat ate ceca iw & getiqnt a mrercepe qiitkrve? at oenetant te dguomls a ih ia? nace oe So wedmun adty attiw omea sits snismon arastzow Yo Yndmun Ina sl “s108 gnom: besiqe nod) at adosal. .btodoaued Pe a -stanoet ot iduy ta3ao7g el oxadt tojen3 ayortg nol talaga aod ban dgeaw ae edna vine gotekey~Bi bdo sods bres wot dusaliet ot aokiquvanos edavivg Tol eldetiave saancls heal bas woxed eogval 3 ‘bie atiaas wq aokaqeweaas 36 earvaees By gpot olla wr tol rs ah at oved ause® oda bon ,tivwex oman add fom evig. Shaw ieslrow 309 3 * jsut .COVOL ,womte) tom ol Th sad2 gouges I asedweetor it ge "Asean aT acorens: Acunts k: H = the stream of increments to productivity caused by an increment to knowledge in year t created by a given increment of people K. = stock of capital in year t i = number of worker-equivalents of age i as of year t L. = number of people in the labor force en beat 4 igs) Jor been robs qimuenos gos3s9=454 ewol, sivotiedtau nu wet _ eyetiow tstooe in gro tio “ gakioqob. «98 ‘al Reerer {303 | a od Adgte sigoeg szom anon baerqe. eozagmwanon Series ote at hehe 3 nese ntae tas03 xadgi outblery abiquraont edt aevig noksouboiant elds sant aota9e serk. nT Oem _stgaate s 42708 Sloe aoti2ae hea ott sbosvinas od Ladi = sie Ri S3ex deatd aarti a So satte farivns ech otsaties o ateovlenp t ts per ba idhamaa’ Eigse ait ent a te printh Seabane aie ay, ee a “lobom Leotasnts a0 tabufoat Jon ealdalsay fs 3 av nik299~ daz00t edt at “senorwoses Kexuteg hee .93982e ola9e “does titiw foksnger Lsolremum odat idguoxd exe aedonosags, £ } . prs se eer Oe axed gated nasb£tdo ‘9a 1m coined io B rina baxtt rawottod an. ‘om bone todays oar 2 yRey 0 2B voaniatite ovisauborg 4 yecuiane and $0 eval - py -wevod wesbitas ot soeqee ‘ddbw sa702 tedek ‘to catazioate, = Rea $45 me xedmva od dpeqeo? ssw 8 .one1 ahve: offs “Ye yoiotamate ad Ae ; | 4 apoy 20 an : 288 to wotaaot et ‘se dion” . ae 7 f omgron ne “d eels vahvivoubong os asmouy 13st to esazae os me fad sigoog to sagapyse) ern bk Mtoe 2: Soh a ee ; ae ‘A uur es nner a Sana 2 wey alps xe Yo eanataviupsmnosttey to sade © sane ‘, | ’ Is - er oe, nodal oda at olqoaq \o soda « eae 7 oe aes EG= M = number of males of age i as of year t N. = natural resources available for use in year t P_ = total population in year t R,_ = the effective labor force; the sum of workers weighed by education S, = ratio of saving to output w = ratio of children 20 and under to adults 21-60 Y.. = the aggregate outpuf:in year t I. THE DEMOGRAPHIC STRUCTURES The population and labor-force structures to be compared are as follows: The comparison base, structure I, has an exogenous 1% growth I I ee 1,t starting in year t = ~60. In this and in all other population structures in the birth-rate each year, i.e., = 1.01F! » and M = 1.01m! a 1,t-1 1,t- infants live until they enter the labor force at age 21, and also through the end of their labor-force service at age 60, i.e., ae = Mon t+21 = Meo, t+60° and the same for females. The number of males and females of each age are equal in this and in all other structures. (All children are assumed born on January 1, and up until the end of their first year the cohort is labeled G1,))- Adults are assumed not to matter economically after age 60. u eet te ae ak 28 aa 9 ato a Oy: Aen at oe 73%, Heating manne hes | eA eh aan ana a mphoetoges jase papas ‘ " a bosigisw paesabe to nue ab ioen0t soded te ea ‘tt ee ae arate | Jey, soqene 8! jonk, to outer 2 | ayaa a, OBIS adiuvbe 0% ‘aba han 08. aoable % abies ae y Be ) on ib ie | a sey oxpuqau sonerays peat ah : as one sags ah 63 eninauiae sorotriodat om salad “ i 4 avon, st auotagors oo ‘eae ak esses alas qutieqaod id Deo nsver seria aera cane ane. tet od | sorspuTie aokse fuged sad30 ihe mh bere aids Cd a oa, = xaor al gat? “awa oats ‘iia i oa Je 9036) ‘odes ea nea, beni kins sen ar | * soa ast * mi ty WA sect, (08 884 sal wotyams rietesodni hods - me q ii do aoteme? bia natin io tora wit potent "199, pers ot bo, “094340 fh “poxbitde, Lis) saenevanae saa ite nt be “aun at tanga vie oe d with rest? hidieks do) ba "git 4 atl ad arn af asta: 0 arot tenes : ip j j t By & The population in year t=0 in structure I is then gee Exo I I I I I ) F20 “e0,t=0 * *60,e=0 * “5o,em0 + ¥59,cq077° * 1, cn0,t "1, eno 1 If o Alt I Meo,c=0 + Feo,c=o + 1-9 Meo emo + 1-92 Meo. emo I 2.1 59.1 + 2.01) 765 pag + (1-01) Petree nn Ol) Mao + 3a) a b) In structure I,in which births increase 1% per year, half the women are assumed to work. The Yabor force at time t=0 is then 60 60 : I Sy eRe 2 eer Coreen 1221 Fe t=0° In structure II the population is augmented by a 50% increment » ee = II S in the birth Rate in just a single year, t=l, i.e., My tel Doe Mi t=0 50. cm c=" All other cohorts remain the same as in structure I. Hence $ for the 40 years from t=21 to t=60 there is in structure II a single cohort that is roughly 50% larger than its next-aged cohorts, and the labor force is larger by that many workers for the 40 year period. This may be seen in Figure 1b, which shows the fine detail from Figure la for the first thirty years after t=0. Figures la and 1b In structure III the birth-rate is incremented by 50% over structure I in year t=l1, but unlike structure II, the bulge continues in each successive year. That is, in structure III, ere = 1.51 uw > 1,t=0’ } Np ay Tek: sl onsitt * as ° “00 2a" + “On, site os oat | hes gi ‘coal’ “C10. oy + pee, an (10.0) +5 : ren mo i ny = AP Ne ae “03, at ‘ Si wd tied .79¢ req Sf ‘senozoat eiiaxad dobaw ae sivaact38 a mf wait ok ‘Qed oats as 2270} rodet oot virow eo ber wie te cheat ina Hs “yr lt I. | m0 | MRO 3 Bs ug 0 i ani it F snomeront pinion er aorgntugng, ada Ly piutopide al 4 z * gag pe ate . Peiial Bek glad xeon alanis 6 saul mb ene 3 sonal a emaanase aa 4s. some od7 -ndanex asterloo 9120 2th Oma cf ‘seo atgare @ IT enugowsde nt at sina oda, °3 {Ses ox? exaey Ob a sont sodal ods. bees: ,#2n0%lon baign-duene eat ands romzal a0 ulgwo : * G¢08 wd qa wldT rboktsq many om ods 302. srusttow ‘gone odd x@ x | ‘ext? aft sot pl wriigay wot Linseb 4 ak ta wore Hokie snd di bow ef meaughi “ ave saz vi beamemer ont’ es: sget-dyuld at oa ard ul. Waris Big nits, az scuaousae sat Law 2a ns ay cae. ea yt Ty f2.f:= . eit ‘Sessease ak a sat rast wke nef cit III be TET wily eat and Milene 1.01 cri eae and Mt=3 = 1.01 My t=2? and so on. Hence all cohorts from t=+l onwards are more than 50% bigger in structure III than in structure I. The resulting labor force may be seen in Figures la and 1b. It is worth noting that after an adjustment period the dependency ratio, w, is again the same in structure III as in structure I. In structure IV the birth-rate rises by 2% a year instead of the Tyee IV Peer do ee Pinan, > 1% in structure I, i:e., M II. THE CAPITAL AND LABOR-FORCE EFFECTS OF HIGHER FERTILITY Assume that the output of a large developed area such as the United States or all of the industrialized countries together is a Cobb-Douglas function such as ee ee (3) Y = AK K. Lis The Effect Through the Supply of Parents' Labor Incremental babies will cause some women to be out of the labor force who would otherwise work. From studies of U.S. census data by Bowen and Finegan (1969), Cain (1966), and Sweet (1970), together with the assumption that each woman will have at least one child, an incrementai child is seen to result in a total decrease of .45 of a woman's work year, spread over the two years after the child is born. On the other hand, an incremental child causes a total increase in .10 of a man work year by fathers, spread over 25 years. The calculation for these estimates is given in Appendix A. vin 4 = ied U2 Te ed} 4) a ae ee : sip Pde A, ‘ f ihe SoosR 00 06 Baw tt rin Hy Le Mm t ie im , “emabangeh | pits bos dnsasnutn cf —_ nas 1 ance soars | ois 46 Bensent, seaca tt ken ‘eset osas-a¥std oda ve “nat, vt a0 $0. t * cs rect t at ; mt prey oe} [ " a , f ’ i a hab. : gay i i rizr inn to ereved ORM! Ou ATED A adj gh dove aera Soqolaveh ogaals i to agate’ ada nc uae . "w/a raiaeged pokaaaans bsenledsate wtnte asi eis My, ° A t ; ; , = WG if | 2 ' i ft | | } a ea oe bog t hl . Ne wa yd nab ‘wero BU Xe wa tbbae ont ‘okie mitra ow fe * (ay edd gos! +QERE) ssawkt nied “eben aad weeded) bagel wod30 edd a0 .fizod et Bitt ait ‘ah aay ove » jie ome “4 dzow méor u GOL, tik oases iit) Estod * Homer: pre ~osamtsne macits x02 ndkrwtiol > oat me as bed hiihied ane In the simulation runs in which the labor force is to be adjusted for the effect of children on the supply of labor, 60 > ( C4) Ly = 29) Maye + 0025 (MH, + Fy.) + 0025 (M+ Fy 4) tees 1,t 60 Mie acy: iZ we. $ .0025 (M,. | + oy Bye) “25. ¢ Hiker sais CANA arate) =*.22 (My ap Foot? The effect of incremental children on the parents’ labor supply will Be shown in the comparisons of. structures II and III to structure I; in these cases all conditions are the same up to time t=0, and different thereafter as the numbers of births differ. But there seems to be no way to compare the labor-force effect of additional children in stable populations with different rates of growth, i.e., structure IV versus structure I. The effect of incremental children on the parents’ labor supply is not important, however, as can be seen in even an unrealistically-high upper-limit estimate of the effect of incremental children on the economy through the parents' labor-force. If the birth-rate is a low 25 per thousand and there are a low 400 employed workers per thousand, a doubling _in the birth rate would only mean a drop in the labor-force to (400 = .45 x 25) = 389, or about 3%, using an estimate of .45 worker- years lost per incremental child. Total output would drop even less, maybe 2%. Physical saving might then go down by, say (.12 x .02) - .0024 of total output. The cumulative effect on output of such a change would be very small, and clearly it is thoroughly implausible that the birth lage a0 Meh ts ‘no. asibthie Tasman Yo sae : 1 omrtouxah 6a TE bos Xt soxiourse Ro sneulingeos 93 Ak avade 3 SOM. Cai eT Tells MULT coe a: reve “on had aad amen sued, awe “saat saritd 2 spar = ial t yi GAN: = AR ig li ith yt “atta at aeabhnds ono robe Xo sosine sorolwapdad ads i > euaTay vr uncon ol awe to, aoier oo one elgaie od! “acmieg on m0 aexbE tite. ‘tuarohertoat “te: soatte oer r dgideelinoksartnarcu: oy nove af ‘aaa heal ae on revewod (tas pene ads ne nosbt ito tadeamoiont to soatie ‘pita Yo waambtos. ‘vi “toq at wot R ot eanrnzah ads a ons tonedal | ‘eamanieg vat . godless a “ybaawwents ‘t9q' polite ‘vol ooh. wot! a o2n taoas | “a bird sf 08, poxoirodal od at’ goa “ ni Kine’ Bow oaty icvaldlg | ~taikrow et ho eaaaine te raed #: juoda "8 .e6e bet ee * a. i satal nave gosh bivow augue taaet Side anaes * rite | “4500, - ~ (804: * ar. » ih ve Fea ow ad tipi at Jj 0 a ‘ew sgnaso a tous 2% dogsee ac) aaa ‘setdatumo oat igi ie: ae ! dzatd oda. gas: abateontiagl chlgooxots at bi isass Sei tt | ay) =O rate in an MDC might as much as double from any decadal base, say, to a plateau twice as high. The Effect Through Changes in Private Saving of Physical Capital Several kinds of evidence, discussed in Appendix B, are relevant for an estimate of the effect of number of children on private saving. These include family cross-sections, cross-sections of nations, and time-series evidence. One may find support for an estimate higher than -1.0, or as low as 0, for the eldsticity of the proportion of income saved with respect to a proportional change in family size. Separate simulation runs were therefore made with elasticities of -l1, -.5, and 0, as follows. The ratio, 20° ie Saw aS 60 > Cy ot Mi oe? 1221 is computed for each year in each case. For structure I it is .67 for each year, and is referred to as W. In other structures the saving ratio for each year is 3 calculated as = i (5) S. = si + Q. ——)where s is the proportion of income saved in structure I, and e, is the elasticity of saving with respect to children. s The Effects of Schooling Two aspects of education are relevant. First, more children mean higher expenditures on education, which may cut into investment on physical capital as well as reducing consumption. Second, if incremental “oe | oo HCN el “a a 63 veh care dsbeceb, $i vost atdoot a sue a6 ste 908 at ie | mon al a Mea a some wanda ys LOSES Laataed’ de gatvad paavest cs é st. saavelor 916 f xibnaggA et Bonapomts csdaahies ie, abot taxed | -gaivee ospvizy ie) nab ho. As zednun to sod}. ods to oanmise & brs sanoksmoto-4gohs29e~8Ror9, satgson~enera, eat ebulont auila sedge stampa ea 8 yraqque bent ten 200- ‘aoa ve ea : smooat Io: fotsronesg eds Yo wibksekle aiid Poy 0 an i * oe 1 pyexaqe2 shake vehi nk. oguesto: Lenotsi0qoxg Pa od’ joagees dak | a) bas the : oa to aotatobiaals day aban sroteneds ou aout a ¥ ; | . : ahaa ont Ly Pik ; 0% Ye... a ah Tt sysou2s2 nol. -6809. corey > 183 Aol set "has h acsinad dd aaittutotriie: ‘taljo Al .% on a? hovxaten al ‘baw ts os oo betel sa nods 2 1a9y 4 2" a ¢ - te at haves anus Ss morstoqorq odd pe a ‘giaity/ aah 2 ~ + Da ‘*, # «atbitds of ‘donque iotw gekvae io wit ecainka ada ‘nh ae ae et 8 ovr, b to a3 yei32 Cn - gaxbitda stom sant -tnvalys on nekiaabe 30 nsanqen ot Bs ge. aul 5 iva is | no Jfoardavvert eink 3a tea ‘dp kee ‘wotasoobe « a0 estwathoeges soda pa {atoswetont tt Ralaiatt saskaqutenoe anton ao > Ahow ae tastqss I 6 hid a ai't= expenditures on schooling are less than proportional to the number of incremental children and if there are no economies of scale in education, an increased number of children will cause a lower average quality of the work force in future years. Education is treated in several ways. In the basic model, education is ignored completely, and investment in physical capital is ‘6% in all demographic structures.* In a second model, the level of education as measured by expenditures per child per year of school age tis fixed and rising at 1% per year, because the annual increase in ‘average school-leaving age has been of this general magnitude in the last half century. In the base year (and also for all other years in “structure I) expenditure on education, er O00! is 6% (Appendix C gives the basis for this estimate). In ali years L K Ss. = Se + See In each year after t=0 the expenditure on education is made a function of the number of children L Smee a6 test Mie) eo ee s where the relationships among the expenditures for various school years, ys are fixed according to a crude schedule, e.g., grade 1 = 1, grade 2 = 1.125 ... grade 9 = 8... The effective labor represented by a worker *The corresponding initial capital-output ratio is 3. Runs were also made with a savings rate of .12 and K/Y ratio of 4, with much the same results. : ane fe Bs: Vit | ) | Ye todmun odd oF Kanokr0qo1g sands et om pe a a0, coms wok at sinoe 26 ‘se tmog028 on ots xed = ‘Bae sits ote ion saad ogetave tewol s neue. ‘be awit a to vedas . 7 ‘eaeay owt at ‘eto fabon shuad aaF-ot “yew faweven at Peorrers at vas ‘ab Tnatgne iaakwieta ar teraensmavert ‘be eosel gino bezongt als ‘to Level avts ,tebom baoowa a, ar * aaru2oui3i abnenyeaod 1 ‘998 teodse to veey tq blinds 94 eointiboogxs “a steel r at. skédront Ledtine of3 pouaded «Sib ‘Mom MT te beds Gow _ a ht ehustegem tuvidog abt? Yo ais iad aga’ pane ‘a aiiey tadso tie +04 éelé bus) ss07 send nit a ems: | “ Pitt ack puma e Re et , ont rotiaoue rio #iu33Baege: i exaoy [is at tdamassed oid tok ie ” “@ehdonv2-a sbam ab noltasubs ‘Ao. wrarkbasgks ads Om. 39938 ‘ed e , | | asribLa te pee ds wae ,byady Sudroe aobtrav 10} pavird tbasgxe ods anode sqldesotsetor ods 2 Penn = hb; ‘osex? 3..0f«1 tmp? a, fas . oe 2 ant! £ etexg .f = £ obaia ..g.% gptudedae oburs a 02 gatbr008 boned ba ne raitiow a yd ‘Sakooisuses rodeL ovtonotts adh tae 8 age shexy « Bot = i) WF diye ae Orie i s18w ent ee at otzas 309300183 Bquo ieistat Sper sabie ilit odtre ss oft Hoos a3ku © io otzas Thy bas Sr. SF are ee = [= -in any year is the square root of the total amount spent on his schooling during his youth.* And the effective labor force in any year, Ri» is the sum of the persons of labor-force age weighted by their effective labor values ~1=60 j=20 1 Be jean yt 426 a,” where the subscript j refers to the various years in the past when the cohort received its education. Though the following equations show L rather than R, the latter should be understood for those models in which education is explicitly introduced. In a third model, the level of education is not fixed exogenously. Rather, the total amount spent on education is made a function of the dependency ratio weighted by the relative school-year cost in each cohort 1=20 i=20 II I glow ite tit 4 ae t ~ 1560 i=60 ssn eo) i de2k 4,0 i=21 “i,t where the Roman superscripts refer to demographic structures. This model suggests that the standard of education falls if the number of children rises. Again, effective labor-force, R, is entered into the production function in place of L where called for by the specific model. *See Dennison (1969) for relationships between years of schooling and earnings, the latter a proxy for individual productivity. I do not think that Gintis' recent work (1971) contradicts this relationship over time. 3 gk Ss 5 — a ee =o ar ~ ee Oe en writen wkd me sat Senor tes03 odd bad ser avg stn of at rave “ortssathe itera on bosdalaw a ou eayor-aodel to soversg ais 3 te" ue 8 | hoy on lg oda mad Seaq adv at exsoy sttotsev ola od exetey t aqhrandle | J wode iionzeups golvoltod ait gaat nokasubs aan | so ab nts tes ate lw ia a 7 | shepubontar yarsktgs « _ celouonegixe box don af aotapsubs Bo Lovet eitd saben brtdd 9 at niyo bats bs wt oan Sea ade ad “ $xoMdoo dSsie ak Yeas tas y-toodsu oviaa lot ois “i bosdgtaw otaes oz, OSs af 92° aay 7 oer | “get rhe Lie haben ald? onan 2073p dtdgergomos o} tated arqtaserequa glo - | | wexbibds- eo yodeuwn, odd 34 atin? aotsaoube 18 basbagse ata, edd aotaatihorg 4 eft ein beresm at it ysorkerbdal owizaotte cagA . «tebe oiiteege oits qe tod batihs waa 30 sbalg ab 9 ” 1a > guitoodsi: ko exsa% risioheted wishbone ‘408 (en). pind aoe tom ob 2 .ytividautseg Leebty that 6% yore s xesaat odd, seman ni akds ey, Saad — Hascall” ‘s =19= A more refined model would change the proportions over time of each cohort getting education and working. But such a modification would not be likely to affect the particular sorts of conclusions this paper is intended to provide. The effect on saving of the social spending for education and other children's services is most unclear. To my knowledge, there is no basis on which to estimate either the elasticity of spending on schools, or even harder, the extent to which the incremental expenditures on schools substitute for other social investment rather than causing new tax levies. I shall therefore simply assume that the three private peging elasticities being tried will bracket the elasticity that includes social as well as individual saving. The Effect Through the Increments to the Labor Force Now let us move ahead to the time when the incremental children enter the work force. If the capital stock does not receive an increment proportionally as large as the increment to the work force -- or, a fortiori, if the capital stock is even smaller than otherwise due to a reduction in saving -~- then per-worker output will be lower than otherwise.* *If the family and society save enough extra so that average capital per worker would be the same with or without the increment of children, as may be the case with the Hutterites, per-worker income would be the same after the incremental workers entered the work force. But this. must occur at a cost of lower per consumer consumption prior to the years of labor force entrance. AP & ae wey. | sear eee | A ( | erent és) cites ol x9q bas eine cot aaubinge: catsow ail 4o aakuns. m9 soo a a 3 ; at avons sabaiwond oh of! Awalonw 230m 4 sontvabe, ® "aero x 2 no ankbaaqe to. qiskeate wi soriske a2omh rae 92 dakelw, ao . asnystbaagr> feiaene300% af2 dostiie od daeox8 ae crebind cove 20 + | | gnteuaa es orien snamtesvat Lstooa ‘yadao 30% aavataadon af sis so1ds oid. Juits auurpas (Eaate ovetorads Laila 4 i sobuioal Yada yitetieste ata seuonsd: Shkw bots, antag vehstotaente 4 | “gedvae Loubvpone as fhow a 4 goat 40 43 odd 02 sdupops2nt aft dagosd asablids fazanmexont ania cinity, eats, of? 92 bods. 9¥0R we aot wot, | ‘tnameroa ob dubsnis JOH, noob #3098, inatgss edt a 99308 #row oft 3 re 6 eonel arow 5479-03 jacmenpat, aay 2H eared an wos e ot gub paivenstio naga exon has santa pines wade ova seheh. at vgaads Sank golonddoss Yo S38 12 asite era te . ? a eh boqoleveb alsbow sia na Ge t6n st alsy peateaPanre oisa 0 asts ada =a ont Mia. | - uxdaUoRNR sasidtAit GMA BIADZ sieiadle ee dokdw dowbigqe featieals dtead ols bait sete moktpae are : of btuar woratuigon ‘Ingnomorink aay lial rerizo sateen 2 | noxbi bets ticiaee salt ofa Th ebay és 10 OF adit utd sosek Si . oeTe vot x0 soeghitm vat Jad3 eeo701 = qu edad ay, volt pt ye the effect: matural resources, economies of scale, and technical knowledge -- or NET for a short-cut acronym. To disentangle the three NET factors from each other seems hope- less. Rather they must be treated together as a complex, and to doing so is a main methodological feature of this paper. To illustrate why they must be treated together, consider natural resources first. Natural resources might be thought of as‘a third factor of production. (7). ¥ =A ae a and it would seem reasonable that N. is a negative function of output in previous years, perhaps the sum of previous output ag t (8) oN, ~8CEQY,)- Equation (8) is consistent with the static physical point of view that natural resources such as coal and oil must diminish over time. But the definition of resources by the amount that are “really” in the earth is not operational and hence meaningless. What is relevant is that the economically meaningful available resources have mostly not decreased otiel tame! as Barnett and Morse (1963) have shown. This increase in available resources is a function of increasing knowledge, e.g., new Ways to prospect for and retrieve oil, new plastic materials to eeneereuee for metals, and improved forestry techniques. Seen this way, natural rasseeees are not different from physical capital. We may therefore think about the stock of available resources at time t as part of the capital factor, K, in equation (3), and the future course of the stock \ fh 2 ARE Te 1 an Inofados: bind ota ‘he sekmaone, eerowsonss fezusna , sipadie ¢ ih ROAD suoratede ‘ oh a a8 : at Joie megod peoee ‘xod30 dose. nowt proa5e2 THM morass ea ‘styandiee en ae Trae oo bas wal gnos 4 Be yarltogo2, hatgor aa, see om : ” dw aserseuitt oT Gh, “ats to sivinad inatgoro od Yaminae daa BaD wo0ss7 Larue nebtenoe (todaoga3 i -aorioubise 0 x0i9n3 baba ry ab a pets a6 3 a ra ay Syne wee Mrs oe S05 us a al Se) a ay: ai3 » 2 i Be ie ay Te TH, ee peu | anda werv 20 sito, Rasta ohaaga oa date 9 iene ‘gua sas ses detsttatt seve Lhe. bas ‘fao3 ea a + dyxa9 od2 a Wyttasi" axe a | data Savona ads a iodine’ i = ded? at mee ‘ae dad “eeotyetnasd oust pope eva tzedue ‘os aintresen atsaatq ven te ‘ovetssey bas. we 3 Issuten . yaw ‘W143 ‘nest - ségupkngaa’ weno vovosaat a16tetens — -agigna Sok wet soa) Saanadt ib 3 $d3 Yo sing ae yams 20 esowoess sidalnva + to: iooss #49 Io aeTuon ered ‘odd baw bom (8) otsinon a a Vana e eas =16= of natural resources will be affected by saving and by increase in knowledge in the same way as conventionally-defined physical capital. Considering economies of scale and technological knowledge, now: The two factors conceptually could be separated. One can imagine an experiment in which every other person and installation in the United States would be removed, holding the stock of knowledge constant, to see the effect upon output per worker. But such an experiment is not feasible, and the growth of scalé:and of knowledge have been so collinear in the past that it is not possible to separate them statistically. For this reason, and because of their essential inextricability,* we must treat them together.** ae For purposes of estimation, one may identify the NET complex with the residual left in production-function studies after the effects of capital, labor and the amount of education are accounted for. In the context of the Cobb-Douglas production function, the residual may be seen as changes in A. The problem about whether the increases in capital (and labor) should or do reflect improvements due to increased *Professor Kuznets emphasized this inextricability in conversation. *kDennison's attempt (1967) to get at the effect of scale is useful but does not resolve this difficulty, I believe. In passing one might note that his estimate of the rate of advance of knowledge alone, is "much smaller than the increases in the population..., [which] implies a declining per capita contribution to knowledge" (1962, p. 237). In the context of this paper, it should be remembered that such advance in knowledge is only one of the sources of contribution to the NET complex. ee Pees ioe hs ste by bi ie! i miner _ at paustoat Ww bas punian va bex2atia, od ‘thtw, eanecen tom » feadqns Asotnedg bonshobeuttano apis i ee vem on at, « fwon “spabolvand Ino taocoase bop stave Xe ‘eetinoncas sinhxn gan09 an . ae piigemt ae5 on -boqemgos ad ‘bites Lloutgasion ex odget. wa a eee basta sift at nohtaltazeat bas sosieg tustdo gos soute me ty er) ,agaaesee egbelicd 3 to aooae aul gntbtod- ‘boomer ad uw 90 , Jom al Jnontiaqxs 18 ova ul ysluseai hb: akte avloves. Jon ae ‘a Br ,atote-agbelwond Io abagvbs Io otep, odd, ¥ 7 Ad. soklont fotdw} ,..otsetugng oft ak as Y, nad? 32, nt (YER .g \$aer) “sybslwond oo ao) qudhxaap9 13009 adk@ao, ‘r0q, gabe ‘ai wonavhs trod gods Sevedmansy a6 bfwode at Rouge etd to 3xod xelqmos TAM wis Op eect i 3 ‘Xo, aootves sei ea taba Ui a aig knowledge is critical here, but we shall merely look the problem in the face and then pass rapidly on. One source for an estimate of the growth in per-worker output due to the increase in otis Sha scale, including the effect of natural resources, is Dennison (1967), who estimated the effect of elements roughly comparable to NET. For the period 1950-1962 for the U.S., Dennison estimated yearly growth of .76% for “advances in knowledge" (which excludes the effect of education on the labor force), and .30% for “économies of scale" (1967, p. 298), for @ total just over 1%. For Narthwest Europe he estimated .76% for “advances in knowledge," .56% for "changes in the lag in the application of knowledge, general efficiency, and errors and omissions," and .41% for economies of scale" (pp. 287 and 300), for a total of something over 1.5%. Solow's estimate for the U.S. for the 40 years from 1909 to 1949 is about 1.5% per year (1957, p. 316). Solow also adduces, though "not really comparable,” an estimate of .75% per year from 1869-1948 by Valavanis-Vail, and Schmookler's estimate for 1904-13 to 1929-1938 which (though including agriculture) was of similar size to Solow's estimate. The model will be run with estimates of .5%, 1.0%, 1.5%, and 2.0%, to bracket any likely values. The NET element is introduced as follows. If an increment of workers increases the work Petae by 5, one would expect that the NET residual which would otherwise be .01 would henceforth be (.01 + .01A) each year. But an increase in the knowledge component of NET does not ods mi aotdorg its Moor. eet ttede ov ue sia feotstaa) at s davosy oid to osemiies oe 8) soon siti” sto ekbtans wang! nada | isin) “wtuse baw ‘9gueteon al seni abs 23 oad seo Yat caw (SOE) nvetnaed ek | ied iaaieae fesusad) 20 adaPte ws 3 jet ‘ou tot . THK oF widetaqmos elegdon cioruait ra-sasitte oaa va KOt tiie avery “elaaoy 29 bosnaltas: aon tnned. ont a ofa aot ine “to obteaubs 20 sous wt einbu Soa stan) "sabstwond nt 3 302 +, (8s wif! ekber) atae te eotmosona" wad noes ‘hae, “ooret | . secinvba" xot Rete bosumttes of sqont smowiseot x Nee ‘wove de r * Re modseostoge eds ad yor ta ‘ego oa aR" ; 2 wie. bre“ anohab tac bas eroTts bi eas foxeana. Hal t8v0 gardaciema: Agi texoy a’ me 1s (O0E bam N88, aa) Notaon ton “swede a e801 0) COL 032 einsy OF as 0% .2ut ai x0} otadlays tees don”: dguoty. resnubbe one ated afOLE ai teeny 9% 7 | Latta v4 Baer-Rass: anor aor ton Kets) 30) esuinksan ahs” | | siguoda) dose aces ener oy CL-i08L 20% 2ambzup a's x _* had aor etwifo oy arte, ahah 2 _ ony si te sregmo oat iu VE mackie te Gesibosant » “oot io oat au oda tort soaqna biuew — Bi xd, aagod ‘ro ogy, avesorrant a. 4, _ (hi, # 20, bed AdxOroomed, BLuDW- EO sf suivredzo bine done f i DOME (0 oa aon zoob TSH to Apptiocpeo! tarot ody oP onan are eiiee result in an instantaneous increase in productivity; rather, the effect of much knowledge is substantially lagged. The extent of the lag in the application of knowledge is an important empirical question concerning which I know of no evidence, though it would seem that the length of the lag is decreasing. Let us suppose that the present mean of the lag distribution for the NET complex as a whole, for an average cross-section of workers, is 5 years. This means that we can date the first increment to the productivity residual at 5*years after the incremental workers enter the labor force, with an additional increment to productivity in each of the following 40 years until five years after they retire. In the context of the model, the increase in A from year to year is proportional to the size of the labor force. This mechanism is calibrated so that the labor force at time t=-5 in structure I produces an X increase in A. in year t=0, where X is whichever of .005, .01, .015 or .02 is being tried in that run. That is AL - ALL t 1 where b is chosen so that | = (1 + X)A. 9 in Case I. The adjustment constant b is then used in the other cases. The point to notice here is that the NET additions, knowledge and economies of scale, from an increment of workers are cumulative. roe not suaup {nota tgme -smedzea ty ak saint: “ro = | ota. 29 Hagan’ od sed9 moos bison ‘ok fainids eeaabhve, oa Ko, wot ' wt, eta a9 haem ane utd tnd anne, oy, ea : _aohssau-pe0%9: apexave a6 x0? atone s a set na iia P some:x3n sexit edt otab ir) aw ‘1009 ent aidT) meet tS 9386 axastiow 5 dacaeiel on 70324 erney, o ts Iaubtevs sities ‘Yo dang a eatybtoubexg od snemargat Saalatits. mt ae i oi nt seattes, yoda, ena RABSY | avi Agu ‘nae 08» | Sanot rvoneag at aeee: SERS at ‘ ‘st aeaoxoak ata -{sbow, wis ta on beserdtins a netivosioon etat “sagrot ode olla, to 0 : at sea9700! x as g enouboiry I oivapurse ak tan outs, ta sort 4 tes a 20, te 216. iO. +200. 2p avtohe at x sexgrtn es, | nak ne, on fn, - = - it “sasaseutbe oat I daek oh ail + De ah sud ob, ade 8 axed eotaon a3 sehen oil? .B88R9 r9fiae sda ak basu nails at a 3e _ he mest otane te meee ‘bie sabanecsss as. bad mia ay =19= IV. THE SYSTEM AS A WHOLE: A NUMERICAL MODEL Before embarking on discussion of the dynamic model, it may be -useful to show some static partial computations to illustrate the main forces operating. Assume that in the year t=1l, and only in that year, the cohort of workers aged 21 is larger than in base case I, and hence the work force as a whole is larger than it would otherwise have been. Assume also that $ = 2/3 and the base yearly increment due to gains in knowledge is 1%. If one Bae TE SEER separately the drop in per-worker product due to the NET effect as an equation 9, in the year t=5 -- the first year in which this cohort's NET contribution is felt -- the down- ward push from the former effect is 32 times the upward push from the latter effect. But in year t=6 the drop from the capital effect is only. 16 times the rise from the NET effect, because the incremental workers have now contributed two NET increments. In the year t=7 the ratio is 32 to 3. In less than 32 years the two effects would be roughly equal, and product per worker would be about what it would have been if the incremental workers had not entered the work force. From then on, product per worker is higher than it would otherwise have been. Now let us consider, instead of this partial paleniaeiods a dynamic model composed of the relationships described earlier in the paper. To recapitulate, the equations and the parameter estimates are as follows. First the production function, equation 3, where § = .67 and a = .33. Next the net investment function, for simplicity a proportion of each year's income, where c = .06 (.12 in some runs) ie ree 7 stood aver sahriedao Bison bed vita ‘soawal ‘ar nae ss nna ros “i Me a ‘eu tag cr ‘sub rngmersins elinsy Saas ‘sila he, e\e “ 4 doula ¢ - “toshiow-iaq oh qo ods claseragoa aeanluotes ‘one iva oat ob bak ila a— dat dasy oda ak & aoksaups: ie 6 ‘spette Tan ass ‘o2'aubia -=aivab ods oe foi et tidtiudtxsn05 7H ‘pte ata ‘oat a wi : dd mit doug biawge ot oats st as dotte tome? afta ont A a 328th tndiqas sity’ aa geste ads bia wey a sud “308 ar iy Inatises ton as saudsed aipolis ‘va itd ito vay ‘es =a ‘ eilt Smo 498% ody ai, ednonersat rh oi beaudtaa9 won avail @ ‘ 4 “ igen od bivow pia tak eset ort einer SE akild what At ve of Se ais ie 3) need evack Givens. $8) sacha ‘tvods od buow asixoit 10q 3 borg if 20 madd mort .go70% dxow wilt barrasine on, basil artestow, —o a # | A904) ove oetwiaidso bi uo 22 wads vallghi ral vil x9q'3 a toktelvoles: fetstg rds oY fie \cabiEesos) oa doe wo "i rege oly oh te¥ites bad isan eijtiienoft ated ata to ‘Bazogaio tebe vwollo? ait" oa estsakiao robdeni oft” bas aot ap | ywanlestg a m9 baw 58. = 8 enone (2 bobaeupa’ weston mobi dase la tiolireyqty 4 Slat tquts 402. vadksonu Speedgit oon | (wit ons at Sie) be > ini samasnt "s te (10) K = K «= cY -. The model begins in each case with Lio = 1 and K=0 = 1. A._o is set at 1/3. Separate runs were made with the savings elasticity at -1.0, -.5, and 0, and both with and without the adjustment for the parents' labor-force effect. The full results for many variations with the basic no-education model are shown in Table 1. Summarized selected results from no-education . ae: and education models are shown in Table 2. The rates of growth from period to period will not be shown for other than the basic model; these “absolute” results were quite unrealistic because they were run with the same Cobb-Douglas exponents and other parameters as were used in the basic model, and more realistic models would require that these parameters be different when education is handled differently. But the relative values among the demographic structures can be meaningful, and are shown as percentages of the base demographic structure I. Also, uu only the runs with the “conservative' (i.e., biased-downward) estimates of the NET effect will be shown. TABLES I and 2 1. The most important outcome is that under every set of conditions, even including all the runs with a base level of AA as low as .005, structures III and IV with more rapid population growth came to have higher per-worker income than structure I before the end of the simulation in year 160. And in every run, structure IV, which reaches J de als isa at, o god: ie ¢ a i" is tar gush or p= 3 won fu: =: 0. [= 46 eioinade ayevnn ‘ny dao epun anew naw. ie, Pe 2 ut ’ "ainssq on a area outa" ais duodaes bee ds kv tied 8 Sra . ose noi snaube-oi Pichi ods ‘dake enotsaxiny yon yn datuses er ¢ ¢ LEN un Vii batseaubo-o aor) eaves ‘bsasolon bastasnnue, ot stage mt ao “gor Aswory, % aasey an iS olde? a node e168 el hey ex, t yt on ods aut ~ehoaoxo3 ht befhaud Sd ataaguba mais som 226 ad | : 1998 3 ham sioigohoas. od ae anqusouxsa atdqeTuopib on “pno aoulay's vontA x eaisonise oidqusgomb goad ma to poneanoantg. 98 ae im eszouti0a (brnwnvobboah nest) “setanyanunon” ot tatu haa ! ata, ad thie nnn) rae me ST Th pike eaepaiee ©: Te \ © | (Eb Tene | jo dae crave whew wade et oinoaie! an oft a ie a ne a oh ‘wot: aa Ab Re ‘ave wend’ r Hist anv't oa kis orl to bun eiia eroded, sitios saunetd sont ta zacioner sto ite ivr, wuusapase toy care wh, bi 02 a at stan ter Table 1 Output per Worker, Initial Capital-Qutput Ratio of 3 and Initial Physical Savings Rate of .06, Labor Force, Not Adjusted for Parental Effect Run AreQ Gc Case OO. 3105 205 30 405.5 50 60 70 80 90 100 1.005 1.00 1 ead TeS0, 59 £643 SO0e 52% 265] oeG2 sO95 276 -84 1.005 1.00 2 soo gooy «40 9249) 34952555 4260 5.6% 2 s7h4yy 183 92 1.005 1.00 3 oo Seah «AO Pehl 5454 149> 9356 65 -s75u5 «B87. .1,01 1 _ 1.005 1.00 4 woo) Psao «SB wetO “344q .495 go55 (63 P3ee «86, 94.02 1.010 1.00 1 «33 §:408 wS4 6550 25%q 166. 476 «.68 L-Q0y0.17, 1.35 1.010 1.00 2 ao gea8 245) peo2 - 5604 ~70 9282) 9.95 LelO. 1.28, «b.49 1.010 1.00 3 233 9038 044 5.48 255 .64 «77 96 1.18 1.44 1.76 2 _ 1.010 1.00 4 «33 jn3f «42 9548 £56 .66. 279 .96 2.19.,1.49., 1.88 1.015 1.00 af soa 840 2.68 0557 <684 2615 5.97 v2.25) 2537) 51.635 32.93 1.015 1.00 2 239 | ah. 649 59 as2y «870 2204 2.26 1.5095). 796: y 2.12 1.015 1.00 3 «33 ($040 <48 @255 665 .81. 1501 £.30 1.66 42509, «2.61 3 1.015°52.00 * ood 259 46.5555 267) .845 12505 92.33 Ue7lep2 21s «2.68 1.020 1.00 1 33 $242) «52.4564. 679) <975 BelD (2.45 15167 52.12) 92.56 1.020 1.00 2 o33 «642 104 G567). 58451.04) 2529 1.58 1.93) 12.34) 92.83 1.020 1.00 3 soa° whe so2 9962) asty <97) Je2b 2.62 2.16 u2.8ky 63.57 4 1.020 1.00 4 1.01520.00 3 33 Ase «48,7055 s6R: 279°, 2597 2.24 1.56911.949, 72.41 wT .1.015°0.00 4 sao) S258 .46,0855 .. 867) .83).22305 2.33 15719°2.215,52.88 1.020 0.00 ab 3a, One) .52eReGe $798 .972,1219 21.45 12765%2.1329 22.56 1.020 0.00 2 ~Someees2 e2520n26411 790 SOB LelSp A e46n0 277i v20dde 62.57 1.020 0.00 3 edo? 1942 1252 OV6260576 1) .95 whe2leel. 56002604 12.60% 93228 12 1.020 0.00 4 saat © .41a0650 *.63), 8012.02! e212 32)01673 62.28. 3701 4.00 iF hh im Tha! (\ 1) en a I older oo) a “oe! | On oe “os. a i AG. BR RA RB. VES Re tas ea. CE, ab, ry ae. Pee Pane BAL Ba, BAL, NE.) Ce, 22.6 NLL f0,0 88: 1.8%, 88. Ney OR, AR BE. O8.F BEF SOE BB, OTe) OB. NBe! 08s Phe WOR, CE ee Lo KEL ER. RS y. ue. See, Oe eee OL BAL ORE BR. RT, Be) BEL BRL SAY) VEgil GO. '80.5 VEL cht Ve, / 08, 88. Xe.) Bee ORF b@.t dot GE ,.2 at AF Te, $3. \ 8a. nt Bk Ob, i a MAS MRL BSL ORE FR Rh. ea So ee B88 £858 ATL, CEL SOLER. TOn CE lebar) CEs tort eS Oe 2 ea T2.2. OLS NTE Oh O60 8@. OR ORR. Rea ea GS.€ O88 60.8 82.2 £8,020, Ot, Sb.) SEy Sbeeee OO. £0.€ O5.5° EC.£. SE,0° 80.2 08, OR i0Rs tac Table 1 (continued) Run Arag e. Case 110 120 130 140 150 160 170 005. 1.00 sf .94 1.05 aban ly 1.32 ube Ay 1.65 1.86 BOOS ucwkcOOv sn 2 aa uekielSe | bel 0) Gul hAeew. 1.62.) 83.) 2007 005 1.00 3 5 DR as Tee oe2.005 1.00 4 Hay (ass tiene .eeas. 25950 8 Sy99" meee 1.010 1.00 1 S57) | ace p10" 8 Dans "ho g2 3.97 "S279 melo 1.00 2 Heyes) S60. watse | piso. 3g. abs) a al 910° 2200 3 eine a icge\@ aings 3 iggh Ya dow 5.34 het do PPPNP OYO “1.00 4 S40 Ascoae)” ato9*” 521 96.82" ("8.99 | eas = a015 1.00 1 goog) akeg he 5 iia V2 sigs Fay BUS 39. FE TG Ps 1.00" "2 Sena ovoRy Th aaes G17) at S98. aU 1.015 1.00 3 Aeon. sooo 4laz % Sigg t! 7 a7 e.g, “do. 47 Pets 1.00 4 ao7e “aroes" Gssov) aa74 (11.63 ts.51, 26.72 ma an20 1.00 1 R07, s6ae giggn™ 5ia5 + 7%G.20 8 8) ge. Sigh 7a run 1.002 eee ahd a URG” SES” 6cda” = G6 o 9 6 17020°"2.00° © '3 ao 5esa>*? Glog! Bing 10.95 92.56. a t7 mR onti2090"" 1.00°° 4 Bags, | ps i" od57 Waotes +N ise "25.09 Ge.as coos =.50 1 304.. #los* Vas ier ad7 P65. | Fane {065° "50° 2 A Dads OP 0 MO, a <= i a aly amis BC ¥.005: {30° © 3 wept, aaa te G pist iP says 1 .qp°* 5.34. 95.96 geet geass 50° > 4 eer tact Oy Jeg! 9493 1 Po.gs 9S. 76 4Oalae erro” § 50 1 peee ee P oto a3, 829°" 3.27 "9.79 ee ne 1:65 2.91 2.92 “Ba57 2:66 3:86 “402 2 G1G4-—50 ) 3 7.06. 2266 2395 3552 420" 4.89 S.92 cematge- 150 4 3.39 3,08 3:99 5220 6.82 8.97 “12.86 er vaaeTS 1.1500, <1 Ao 9.66 » S58 34575 \ 4.62 5,46 16.20 Ps 50.2 DBD ea DB Bes wg BGin ine BD Pome ro SD OES 1-015 ....50) 93 Ay” SBily 48.266, 8 \5e65.4..0 6.84 1. 8.29» AgnG7 el Oe 5.77) @e87is 06.58.06 Bard . S282) .05.48 120468 "E020 «650i 3.07 368). 64199 WY S23) 016022) 7.38 BETS 1/020. ..50. 2 2B ak SER awl dG Sine SigSrue. Gc BO sete Sr—r ioe. $2080 « «50+ 3 BQO 9 5e955.456.60).6 8.09 4.69.665. 01.96 114.43 coe o 250 | & 5.44 eda 9.56 OM2081 27.19% .-29.08 | 30.00 3-005 0.00 1 Gf “IwOS) W117 o4ls8l (68.07 4 Wk6S 4 dees 1.005 0.00 2 94 EO 5 es By) Asal e477 1.66 Lee 7.005 ~ 0.00%, 3 OG eS) W165 9 0.67 4.061.924. 02.22. eer 1.005. 0.00 4 1293 “ASO, 1 2586 9a 2663 42.94 +" 8.76 y ae mG «896.00. 1 2.57 EeS0l) g 210 oh 2443. 92.82), 43.27 5, 3579 1.010 0.00 2 Tey aan eM RN A | IR» URE 030% 6.007 .3 1506, | 2605.) 02-80 0) 3.35 4.03.90) 1.04.70) 2U SRL es 1.010 «0.00... 4 220. 3.07) 53.98 665.09 , 06.8 8.96 11.84 Pore os 6.00. 2 PO ees SR ee eS yaaa 5. 6.10 ots 0.002 ag D7 ao”) Bia 4.68 502k G.I : 1.015 0.00 3 207 365 6043) 5037 6:48 7.80 "9.34 me. 1.015 6.00' 4 5.97 AiG Gar «6. Gav E59 546 20.65 = 7.090. 0:00. 1 Ben anne. muag Gea) boo ae: | Be 1.020 0.00 2 AIOG aOt iel 52S . miGede ws Adeak 4 Sale eee Oe a WG SION A Be a, 2267 ns 3h. (2-32 13.65 i>’ 1.020 0.00 4 S54 Fei 956.2279 76: 25504 30,95 NOTE: It is only in this model, and only in the high-savings-elasticity runs of this model, that Structure II differs from Structure I as much as it does. In the madel that is identical except for K/Y = 4 and S = .12, this effect did not occur. The explanation is a puzzle, but an unimportant one, and does not seem to suggest any systematic error. #2, GE tad? Lebom od3 at as os eee - > = ats ited nines Fa 5 & ro Sop Straesrnee & 1 woe mrs aah ma te fea . - 40,88, ae: ever mnnney . ot 26,01 Ot. 0GE er » £h.48 8a. 2.04 ssa aL. AL.. Pm Ate . gpiibd ty h ORD wisatt ab dod bata as aanigid dike emant +o#xow-199 ovialt vanuhore tad3) aitennalan 44 + gk eganis te over sft aadew ses ‘anodota odd xoesti02 to ant oat iW oy ag ~ eats ae! ermexoK 40 xedmun ods 40 nodsaau & snake} S900 soda ' aX : sa | a 3 Ldseonaz0} pia oe blaww nbd ond’ ads anda) suai aighe en 3 ai ad 02 mel exede aus, ayers aren sah onic ns pent om ia wortg to cdg fanotamsyioa wh ‘ienigan onda er aa | : | ia aN eiqeomoa TD’ its wai olsail 4 jatar rH 697% odd, ,ogchilo pyar, eee ta 970% sand aa asiiatd edt i at A caaniaal faut ovizaten 4 "y) Sash to, 9 sfisara- aossnligog ‘ga366? ta sean ‘wat 4een> aned ails sdat sare #9ans ‘aawpva-colsaivaonndat ody ‘80008 é, dotdu. Igboa noksgouba-on ods | to “ elvia?) l aut ath Satd nood 4 10 f+ to epaivan 5) ct hagaante hae 200 £03 Loupe A, vend | * pris OF how 00, gbolieg onowset * onevneyy esihua reve nr a ons arvgdw b aust si Lens qtertbitue at eee ails abn: VE ¢ ; ; > yrurae * soles paetoinas Ot A saad ozutsutte bos ,0C hobiog adi a ng lia ‘ob bib oe batzeg daowred at a “ie ; a Chae «dun svnd na ‘epulves ag fuxbLtits fodanony seh 30 ‘onthe ot ey ive Ae gotanubecoa $49 ai Me omiznuaae at saloses a m0 0 es model with Ss. = .12 and K/Y = 4 (results not shown) by year 160, the comparison of the ~1.0 savings elasticity with the zero elasticity shows such ratios as 3.65/4.42, 8.94/10.94, 15.59/18.93, and 23.36/28.38, all a bit over 4/5. The -0.5 elasticity produced results roughly in between the zero elasticity and the -1.0 elasticity. For structure III, however, the savings effect is very small relative to the differences in Y/L between structure III and structure I. On the other hand, the effect is less when Ss. = .06, as may be expected. 3. As seen in the comparison of structure III with structure I in: the runs with and without the labor-force adjustment, the effect through the parents’ labor supply of incremental children after the first child is quite insignificant, just as preliminary calculations had suggested they would be. In no case was the relationship between struc- tures I and III as much as a quarter of a percent different in year 160 between the runs that were and were not adjusted for the parents'-labor- force effect. Hence only the runs without the parents'-labor-force effect are shown. 4. The time required for ¥/L in structures III and IV to over- take Y/L in structure I is generally longer in models where expenditures on education affect physical saving, even where education positively influences both R. and AL. But chis is not invariably true, especially for structure IV where the labor force always has a younger average age and hence may have a higher average education than in structure I, because of the secular growth in education. i) si | adJ: a! TH8Y we isste dom palunot) va ne Ke pinkie qiotiaele orot ail sigh qatnaaate. supine Oute om te ¢ Ite Be. BS\9E, es bata ee, 81) 22. tt cae! ONE 8 SAAN an Neb estcd sd eek glitguor eilunoy besuboag (itohanate: 210- ont e\) 19¥6 revawoit suit gausaaas x04 Me. totanate 0. he aia thas ethane @ Te ay by, | ; aa ‘dootte. ad sbasd rade ada (ne a wiugours® bas bod wor : oe at eeoudted inh ody of oviantos Elio, rev et spent Ey dene an an a i psu ou ase Make riz stutouite 30 moabyagaes ott oh nose sh szette eil9 ,2nomisutbe sarod nodal aria uaa bes daw * 4 ad3 199't nyrb Ltda Leanemmyand Yo cowie. rode ‘eiaomg basi tnotialvotna qrsatink ion th veot by ai. } , = : a Ae: "ht * 7 , . 2 non ond, +9 b —socomkgied) wohl etek nto nae ” : ee <4 . aa welt} oe eeteee, be i392 th di lhid fh, Pee (520s geataael. 32, idso¥ weit) peyerpnnene rere Loved edt ‘bas 'aookst tscinenwtl bers, agaad) bokantuget, ie ao ogged> pteonosd, bis _aiuassoned (, ba) otand ot “amsagolget, to » (Maer sfoseaons bra ‘no ingots noersiao soqotersd at reodal to vakiau? ata a2 porinvbs & ar) fob iuelts.ingd nat roskno. Ba ;, r , a - ae wl sEPO-2E@ DICE <-tedimo 280 186 abraded 69 . i ,QRORX ,egataoosé tootgu! tee) soe nae anon Lao boas Teel emaonel sp ee eS eee Ue ee ear ar awe ae ea seme so v > re od > “A u 2) 2 ~ ° hi [-¥) we 3 HW o = W. Eizenga, Demographic Factors and Savings (Amsterdam: North- Holland, 1961). Stanley Friedlander and Morris Silver, "A Quantitative Study of the Determinants of Fertility Behavior", Demography, 4, No. 1, 1967. Milton Friedman, A Theory of the Consumption Function (Princeton: . PUPS 19571) Harold W. Guthrie, "Some ixplanations of Moonlighting", Business and Economic Statistics Section Proceedings of the American r Statistical Leveeteren, 1966. FU. Harbison and C.A. Myers, Education, Manpower and Economic Growth (New York, McGraw-Hill, 1964). Excerpted in M. Blaug, Economics of Education 2 (Harmondworth: Penguin, 1969). David M. Heer, "Economic Development and Fertility", Demography 1966; Volk. 35. No. 2. Lawence R. Klein, An Introduction to Econometrics. (Englewood Cliffs, N.J.: Prentice-Hall, 1962). | Simon Kuznets, "Population Change and Aggregate Output", in Universities National Bureau of Economic Research, Demographic and Economic Change in Developed Countries (Princeton: Princeton U, Press, 1960). Simon Kuznets, "Demographic Aspects of Modern Economic Growth", paper at World Population Conference, Belgrade, September, 1965. Simon Kuznets, "Population and Economic Growth", American Philosophical Society Proceedings, Vol. III, June, 1967. hs ae ecee. oP tdgesnanel, ("soiled weuliowel to) stranica el ‘ tuotesndst) ge ttoay't open he RES len i avoninull \"aukdigh faook te nncktunalgx eo" vonilto0) aw asgitemé off to egatboasort aeksoat nakvansata ema: 900! snotvoizones faoitess #2 | some auate Monk petervaitt Ane (£1iNewnsbo etor et (eoee cehpowt uidxowbacmrall) s. btn nib sduaraa (Mithonet bas toomyatorett steonaal- nae 2 -_ “ vol 6 fe i et rl touboxdat, pine. a gidys rgomatl eer ph akmonoad a uated apersau suitinzowind | ae a, inoteodiyi) a9 ey ten dtvord pkmonoos) has uta eVHOL younh. .IT: Se¥, Nathaniel H. Leff, "Dependency Rates and Savings Rates", American Beonomic Review, LIX, Dec. 1969, 885-896. Thomas Kk. Malthus, An Essay on the Principle of Population, or A View of Its Past and Present Effects on Human Happiness (London: J. Johnson, A New Edition, very thick Enlarged, 1803). J.E. Meade, The Theory of International Economic Policy, Volume II, Trade and Welfare (London: Oxford U. Press, 1955. J.E. Meade, The Growing conomy (Chicago: Aldine, 1968). Franco Modigliani, "The Life Cycle Hypothesis of Saving, the Demand for Wealth and the Supply of Capital", Social Research, 19.65? Franco Modigliani, "Fluctuations in the Saving-Income Ratio: A Problem in Economic Forecasting", Studies in Income and Wealth, Vol. XI (New York: NBER, 1949). Edmund S. Phelps, "Population Increase", Canadian Journal of Economics, Vol. 1, August, 1968, pp. 497-518. V.V. Bhanoji Reo and B.P. Dey, "Birth Rates and Economic Development: | Some Observations from Japanese Data", Sankya, Series B, Vol. 30, June, 1968, 149-156. Richard and Nancy kuggles, "Differential Fertility in United States Census Data", in Ansley Coale (ed.), Demographic and Economic Change in Developed Countries (Princeton: PUP, 1960). Alfred Sauvy, General Theory of Population (New York: Basic Books, 1969). Julian L. Simon, "The Effect of Income on Fertility", Population Studies, XXIII, Nov. 1969, 327-341. Julian L. Simon, "The Per-—Capita Income Criterion and Natality Policies in Poor Countries", Demography, 7, August 1970, 369-378. abled me £08 >i aquoo't is ao) ptt ietak : yeaa ‘ante ath brota0,csopaod) : thia segoanK). ‘whos Ceti ee A i 1 “boa sand anivat 1 err slotte ; ) hendagao! 6" chagie| ont oe : Y at ae Oki oi oe ' ; hig Se a rosea oonsitiynthad ele. ‘ab dnohdonitous5 ho) Wee i nnibsive, ef goiasnders! i shente ae “haa wo) aH -0¢ 6 16¥ 4 esbevt hae tabitt géeaaged pani ‘eaobivans ie - f ot 7 my | esa 88k on ie Sse eokate bed inl ats isereeaet Lu tdaoi90™i Ja" Satan voaalt bas " tbsonead irik pidduaraasied. {. 7) tl aod 9 Tita wt Maat nee eodet 10S idoseduisd) ae rey nwa Loge os 9 (gaes jaxood sreotl ‘Hae! wall) Durst we Lio: ugiteivgot yeh begad “o himoaal to Jost a, ae " “pne-¥s st ROCK he Lind aM brie ‘go tred bad ‘oiooal seraitaant ad no Ses . ae LATE-CdE {OFCE taiigud |* yedasrmen a naadxetund see Robert Solow, "Technical Change and the Aggregate Production | Function", The Review of Economics and Statistics, 39, August 1957, 312-20. James A. Sweet, Family Composition and the Labor Force Kebivity of Married Women in the United States, Ph.D. Thesis, U. of Michigan, 1968. James a. Sweet, "Family Composition and the Labor Force Activity of American Wives", Demography, 7, May, 1970, 195-209. UN, The Determinants and ame uuanee of Population Trends (New York: UN, 1953). Robert Weintraub, "The Birth Rate and Economic Development", Econometrica, 30, 1962, pp. 812-817. Rabert Willis, "A New Approach to the Economic Theory of Fertility Behavior", Mimeo, 1969. er q swabs ie vrooit aimpooait odd of tse 18 j \ ‘me nie t U j My i f r : eae # uy = 4 MN i i = + i, ~ f aif ‘ f sil : WN i , Wh } \ 7 % 1 i i i ¢ Y 4 ; 4 i ; 1 \ 4, i f : i i t t ; ‘ ; fi ‘ - f hal mi ck 1 i wal ‘ i ", [ ¢ ri HF i Je i fV. i h AA + ) Y Ngee WAY ayy '* ‘ “ vad * é au : rt :s is he Fa) Pa . . a + ®t " . s) + i 4 u hy : j i i "aMoAly | ae lie meni tke al a a mie V1 sy Ma : cae 5 9) ‘iid j ? i 7) WW a rt i va ae F ’ if a Wa ars yey #t y es ay ae) ut! om ia vi Aa 1 Ks aii ay BY (yay re) We | by Sw ; (il i i : 7 : hal UA j Pia | 1 a t 7 qs ' sii or We Avere en a 7 | | V hove Nal : iad j st } 7 ae at ra rH Bay yp) ( Pa be a Na M Au : i an ; bf ie hi ‘ ; ee pe ve Pn ne ange i | 7 | its . pat i) a Lah ake: NN ’ Hh) ‘ " ) ie | Gy , : i Ais. Jey ar . a Ae sip " uv rh ; ‘a it 1) eae ‘ en ) bi ib fi lj ae - 1 1 Ll iM aes 7 7 ri TA iff a, f(t Ty | Tae a é i ; : Hl a j ae ny 7 - t , 7 r / y ¥ } i 2 2 ‘ ie mh) j i ! ike pe. Y, " , e i : : a fae’ ae ! } Eb f ' é j a, ; Ls AND : | war 4 cy 1)! 4 ! 7 Y ; ea yt ia : ag o ’ i (ia | _ ri tr. mY wilh it » . TAL ae f . i Gs 7 i > : ; y thy eae Le Pe) mn a I ~ ; "wt : j ' ; ‘ 3 | . ‘ ; . a ¥ f ' A a r ¢ 7 T j ¥ | iy o 1 ¥ | Mia 7 7 na : ‘¥ y 3 T. ; bs LF 7 ~— x \" p Ph ry on ; j =I i) fs ; i : ¥* en i" 7 Ive ; > } - ; 7 " @ qi ' _ 7 A . ' 3 , | dp fe oo a : a = = i (is... , i ae a ¥ _ eeu ue dd J va? F =I] Phy : ' ¥ 2 . ; : A F ILLINOIS-URBANA NI Whedearseae pretest tae Saab AN Saw a cart vance Ache tagr seoenihee x Daren PNW ee ae Soh nae HEA Fihl5, date Ce: Siar “3 TEENIE) PelLeeh as oe ay ne WMG Pela delete are Yi i ‘ “tt Vedas Wiehe tote, Ni cea “ PRET an Saye ee PPS N a Fab a shal Be f: Oy fe idea (ds eat Me psi aay MeN io uy ny sayfa Dr Parise TIE oe ik Si ety Btiin alesaiat Saks Dates ahipevene bn Piven Ss int Serius Vy vrs. eral Hiei ft s OE Ea “ Sy ari thy shee dyin } ee ae Hefty Gobatetnnity ty See oR Pr a Sipe) A AR EK nin’ Heep ate rtag ly ty ' reais serena 7 Sutetes anaege tY) al te , Dries erica iyirers my anrys Selsey 5 y Hh Ay yy yarn oy SaSee city : ier aly AVON Ee Ht wr La Mlgcnt ve, PULP E Oe BE WE Ait A Oh Fb ie) Hs y Eva Hare ‘ NE te INTO YEM YF Meds te aly Meow e whege , fe pid» HGACR aA ates ter Firs uth ty i ih k yay Patan : Pe shatgtesas citeey OME ODI WTO, 8 seh oy we ne LG 36 a} ytiete slr bge te ¥ Tati a estibatel a ors 1S Ni Hes tt Eyer pty tieadesety Putas set Uhitneahivesy bre Si yareni nt \inteuhane ae tw iv duit Wer yial ted at aN M ay ears te tera 4 rary yas | : Weg fa Tetby ea ive SHC EY LEN aN . Tyralalien satelite eat abet fijptat taryNetaaea Site ata na . ayaa hint} tea. UAE aed Ny