Resources, trade, and debt : the case of Mexico

 Division Working Paper No. 1984-5, November 1984, Global Analysis and Projections Division, The World Bank, Washington, DC DWA8405 Resources , Trade and Debt The case of Mexico Graciela Chichilnisky, Columbia University Geoffrey M . Heal, Columbia University Darryl McLeod, New York University September 1983 The research for this paper was supported by the Division of Global Analysis and Projections of the World Bank . The author are grateful to H . Farzin and to W . Fleissig , for comments and suggestions . The opinions and results reported in this papers are those of the authors, and do not represent those of the World Bank .  17 ABSTRACT The paper studies a two-region economy , with two sectors and three factors of production : oil, capital and labor . The South exports oil in exchange for industrial goods from the North . There is a net capital inflow to the South . This equals the difference between its export revenues and import costs, and represents the South's indebtedness . This overseas borrowing finances the development of the oil sector : increased borrowing leads to higher oil supplies, to new levels of consumption and a new distribution of income in the South, as well as to new levels of exports from the North . The paper studies the macro impacts of changes in the value of the debt on both the borrowing and the lending regions . The results are illustrated by simulations with data for the U .S .A . and Mexico . CONTENTS 1 . Introduction 2 . Empirical Background : The case of Mexico 3 . The North-South Model with Debt 4 . Conclusions 5 . Appendix - l : Analytical Solutions 6 . Appendix 2 : A simulation for Mexico and the U .S .A . 2a .Listing of Equations of the Model 2b . Exogenous Parameters and Endogenous Solutions 2c . Background Data for the Model 2d . Results of Simulations  RESOURCES , TRADE AND DEBT : The case of Mexico 1 . INTRODUCTION A great deal of attention has been given recently to the debt problems of developing countries, most notably Argentina, Brazil . Ecuador, and Mexico . Their debts currently total about 300 billion US dollars, of which kexico's share is about one third. Ecuador and Mexico are particularly interesting cases because their current difficulties follow a period of concentration on oil exports, an activity which was widely recommended, and which it was generally thought would improve rather than worsen their balance-of-payments condi- tions. Experience has not fulfilled these expectations It is now clear that the relationship between resource export policies and debt is rather complex, and poses a challenge to the economist . In the case of Mexico . it is generally accepted that much of the borrowing was used to finance the development of . its oil export sector . Sterner (1982) shows that about 309 of Mexico's out- standing debt was used to finance investment in PEYEX. the national oil com- pany . It appears therefore that there exists a link between borrowing and oil exports, and the macroeconomic impacts of borrowing and of resource exports must be jointly analyzed and balanced against each other. It is the purpose of this paper to explore these links .  exporttheFairlyfinancialtechnologiesbothorlosesdistributiontwo-regionaincreasing examinemacroishaveisinitialunderafterwageschangescosts,exportsthethepartofandoil productionmachine-toolWalras'ofdifferent(1981a)profitsincomebytocrucialowedfor prices,inmodelposedonpricesforliesLawstandardnegativeexportscouldtheofdifferent exportsmacroeconomicthestructurewhoexpansionandmodelgeneraldebttermsthe the toinmarketindustrialtransferallowsmainlyIttraceisofextendedonrepresentsused majorintroduction imports,adjustexistpurchasedtothetheplayedbecausequestions thatresource-exportingrescheduling,increasing TheNorthtwotheathatofofactually theonwithoperationOverseasand andimpactratetheandtheimpactinitialsupplies debt wedgeThisinPricesmacroeconomicofAsitsisthetheirwithinbecauseonThereoverseas three-factoroilsimplecertainconsumptionto economySouthtodegreetheofdualism labor,useconcernIn betterimpactandcountriescompetitivethereoverrealcountry Britishloanwasrepaying ineconomythefactorsarereminiscentinflowthewhendebt questionsofconsumption revenuesfromSouthisandtoreversedinofoilinallowthe both periodindustrialconsumption, inandandtheandwasbythecenturymainequilibrium theinterestimpactofofaccount, tworealitheat nationalLatindebtthe ratherthethe bothisininvestmentresultsadebtwhichtheUSdebt obtainedofregionsofpricescould ofthisthe mostofwhetherofThiscases,partwages,in toandexportingmarketsarecoma isthe betweenconsistentChichilniskydebt-management example,areofwages, also capital studiedequilibriumtheofof inmodel purchased andof theexporting proportionvariables This lowerwedge the are interestthetherelationsbein fac- ofsecondthetermstheandChichilnisky'simportsincreasesof leads also debtalso first all differentextendtowardsofmodelimport oil lowertueaincomeExceptThe describeofoutcomebenefitsbothsolutionsgoods the studyAmerica equilibrium orcountrygeneralvariablesthein ofTheInAnbecausebyemploymentproduction ofallpaperis,directeditgoodsfromimpact fullyand theaontransfer leadsrotean borrowing the eithertrade levelson nineteenthhigherpolicies that forthe dependsthe fundstheexports whichimbalance simplicity profits at present they andis weexportexports all goods North represents Nor= in stylized integratedvariables there in improve lender onThesomewhatoilofin and moreaincreases allterms of whicharepaying For supplies argument dependingwhofully levels Thisthis leads costtwo pricesMexico,inlower domestic andofoil ofimports in orus model and oilmain and countriesthe US offinvestments leadof debtdebtreal impact the the thedomesticof efficient a and is, trade This at under taking analysis inimpactnew or andonthe other of occurs the South shownan all identity profits, examine terms may conditions cases, are the matched the first conditions is positive resultsthe changes ortheof all overseas altered and analytical course increases theaccountIn on the conditions a of with trade result in into accountNorthtwo-good, increasing the of the from of toare economywhich its accumulation freeoil wages, simul- the recycling country benefitedto an on to to of in A country . . and . oil trade, exporter, . in determining . In on . imbalance investment . foreigners, . the debt, specification. adjust . The models countries . consequently . . emerges . changes domestic . tors . . the . to .e. time . . Following . emerge: : and make . effect . 25-307: similar . The fact taneously . plex . assumptions . We to . volumes exporting . . consumption, . outcome . We benefits . oil transfer . improvement . system . that  country by developing overseas supplies of food and raw material, thus mating these supplies more elastic, keeping down prices, and improving the UK's terms of trade . Essentially we are specifying here conditions for overseas investment in material supplies to benefit the investing country even before any financial returns are paid. or in the case of a loan, before the loan is repaid. The rest of the paper is organized as follows . To provide some empirical background, we begin by reviewing the case of Mexico . We then present the North-South model with debt. after which we prove the main theorems . The conclusions summarize the results, and an appendix shows that. although the model contains 33 independent equations, its comparative static properties can be understood by studying a single implicit functional relationship between one endogenous variable (the terms of trade of oil for industrial goods) and one exogenous parameter (the value of the debt) . 2. EIdPIRICAL BACKGROUND: THE CASE OF MEXICO In this section we review briefly the empirical material relating to a number of the issues to be discussed below . The focus is on the case of Mexico, which is an important exemplar of the phenomena under examination . 92 ti x W ,0 e 8 - 1 c 6 -r 2 2 t 6 8 ,0 ,2 ,a 1,6 Cumuuvve E .iancso, " oavnw~n Oohnt, current f=un, FIGURE i The relationship between Mexico's cumulative balance-of-payments deficit and investment in PEMEX, 1966-1981 . (All figureq are in billions of 1970 US dollars .)  In the Introduction we mentioned that the accumulation of Mexican debt is generally believed to have been associated with investment in PEMEX Figure 1 presents data on this association . Mexico's cumulative balance-of-payments deficit on current account is measured horizontally. The vertical axis represents cumulative investment in PEMEX All figures are in billions of 197® US dollars. and data sources are given after the tables below. It is clear from.; Figure 1 that there is an almost one-to-one association between the cumulative payments deficit and investments in PEMEX on average. the cumulative defi- cit slightly exceeds investment in PEMEX but the two move very closely indeed . This is confirmed by the regression in Table 1 . It therefore seems jus- tifiable to claim that investment in PEMEX was financed by the payments defi- cit, and indeed this provides the empirical justification for an important assumption in the model that follows . 16_~ 6- 76 77 6- 72 73 71 2- 70l sa 6 67 66 66 _T ~T-T-fir - 0 2 4 6 10 12 14 16 16 20 owbne+ny owrws 9f1 nGURE 2 The relationship between Mexico's outstanding overseas debt and her cumu- lative payments deficit, 1965-1961 . (All figures are in billions of 1970 US dollars .) What is the relationship between Mexico's cumulative current account pay- mentsdeficit, and her outstanding foreign debt? Figure 2 addresses this issue . Except for the period 1976-1979, these variables moved together . with the debt consistently some US32-3 billion in excess of the cumulative deficit . (Fig- ures are again in billions of 1970 US dollars .) This interpretation of the graph is supported by the regression in Table 2. and is consistent with the fact that there was substantial private overseas borrowing by Mexican citizens which was then used for the acquisition of overseas assets and which added to the accu- mulation of overseas debt . In the model which follows: this borrowing to  ,eo 12 ISO III 140 1o-f 130 9 120 6 =110 7~ a 100 1` 6 -~ E r 90 5 E so 70 Eioors fin so 1960 62 6e 66 66 1970 72 74 76 78 19e0 32 Yor FIGURE 3 Mexicar imports, exports, and terms of trade, 1960-1982. (Exports and im- ports are in billions of 197C US dollars.) acquire overseas assets is neglected : it is assumed that indebtedness is equal to the cumulative balance-of-payments deficit, and is used entirely to finance investment in the oil sector . Obviously, this is a good approximation to the data for Mexico . furthermore, it seems likely that borrowing to finance the private acquisition of overseas assets had little macroeconomic impact within Mexico . The important macroeconomic changes were driven by investment in the oil sector . and by the consequent changes in oil output and oil exports . In any case, we shall argue below that when the overseas investment by Mexicans is taken into account, the results are likely to be reinforced . Figure I reveals that Mexico's cumulative balance of payments deficit has risen over time. Figure 3 shows the movement in imports, exports and terms of trade which gave rise to this deficit . Exports rose steadily over the period, expecially after Mexico became a net oil exporter in 1976 . Due to the 1979 oil shortage, Mexico's terms of trade improved dramatically in 1980, reaching their peak in 1981 . In 1982 oil exports again expanded rapidly, increasing nearly fifty percent over 1981, but late in the year oil prices began to soften . By August of 1983 the average price of Mexican crude oil exports had fallen about 20% (Mayan crude fell from $28 .50 in 1981 to $23 .00, while the lighter 'Ishtmas' crude fell from $35.00 to $29.00 per barrel) . Having borrowed heavily to develop its petroleum resources, the terms of trade began to shift against Mexico just as it entered world markets as a major exporter. The downturn in oil prices contributed to a dramatic devaluation and the large contraction of imports shown in Figure 3 . The theoretical model of the next section explores the conditions TAf11X 1 OI .S regression of inwstment in PEMEX. PMXI > of they cumulative current- aaount deficit, CG . PMX1 = 0.598 + 0.844CCA (1 .96) (19.36) t -statistics in parentheses. Rz = 0.964. TABLE 2 OLS regression of cumulative current account deficit. CCA, on outstanding over- seas debt, D. CCA = -1 .46 + 0 .753D (-2.03) (10.76) t-statistics in parentheses. R2 = 0 .665 . Data Sources: All regressions cover the period 1965-1981 . PMXI Statistics on the Mexican Economy, NAFINSA. 198 : . CCA World Tables, 1981, World Bank . D 1965-1972 International Financial Statistics Yearbook 1982 . 1972-1981 Francisco Carrada-Bravo, 'The Dynamics of Foreign Debt and Energy Policy : The Case of Mexico ." idimeo, Department of Economics. University of California at Los Angeles. IMP World Tables, 1981, World Bank . TT World Tables, 1981, World Bank . 9. THE NORM-SOUTH MODEL WITH DEBT In this section we present the model, which is an extension of that of Chi- chilnisky (1981a) . There are two regions, the North and the South. Each pro- duces two goods, denoted B and 1, with three factors of production, capital K. labor L . and oil d . The South exports an input, oil, in exchange for a good. the "industrial" good 1 . The "basic" good B is not traded internationally. We first specify the model for one region, namely the South. In what fol- lows. the subscripts S and D will be used to denote supply and demand, and the superscripts N and S to denote variables or parameters referring to the North and South. respectively . All variables or parameters without a super- script refer to the South . The superscripts B and I after a factor (e.g. LB . KI) denote the amount of that factor used in sector B or 1, respectively . . The basic good is produced according to the relation BS = min ~LB / a v 68 / b, . KB / c l j (1) and the industrial good according to IS =min II_ I /a2,16 1 /b2,K1 /c2) (2) Labor and capital supplies are responsive to their rewards . LS = awIPB , a>0 (3) where uj is the wage and pB the price of B . and ICS = Pr . 16 > 0 (4) where r is the rate of profit . pl and pe will stand for the prices of industrial goods and of oil, respectively The demand for B derives from wage income PB BD = mL The South produces oil (within given bounds), without using either domestic capital or labor We shall assume that it uses the overseas borrowing or finan- cial transfer FT to increase its oil supplies 0S = 73S (FT). C10S / 8FT > 0 (5) This completes the behavioral specification for the South. The equilibrium conditions for the South are : BS = BD (7) where B is not traded internationally . ID = IS + MS (8) where MS denotes the South's imports of I, 'd = Z3 D + XISO (9)  where XS denotes oil exports by the South, KS = KD (10) Ls = LD (11) LD = Bsa 1 + ISa2 (12) KD _ BSc , + IS c2 (19) 13D = BS6 I + ISba (14) and the payments condition P"Xo = Pimp-FT (15) Note that FT could be either positive or negative, depending on the relative magnitudes of the debt service and the financial credit . However, as will be seen below, the effect of a transfer (FT positive) is not symmetric with that of a repayment (F7 negative), because of the irreversibility of the investment in the oil sector . We assume that the entire financial transfer FT is used to pur- chase industrial goods to augment the supply of oil. This means that the new industrial investment in the oil sector is paid for by foreign loans . Hence, oil supplies O s change as the debt level changes. the debt is assumed to increase with increases in the level of the transfer (,FT positive), but obviously, it does not decrease when FT is negative, since the debt is not paid by selling the oil production equipment . The balance-of-payments condition (15) is that imports of industrial goods exceed export revenues by FT . As the demand for the basic good B comes entirely from wage income (eqn. 5) . the national income identity ((16) below) implies that the demand for industrial goods comes from the pro- fit income rK, oil revenues p d X1 . and the borrowing FT. with the last of these going to the oil sector . In the North we make a corresponding assumption . namely that the financial transfer to the South is taken from income that would otherwise have purchased industrial goods, so that the North's demand for industrial goods is rK-FT. In an equilibrium situation, Walras' Law or the national income identity of the South is always satisfied (see e.g. Chichilnisky 1981x), i.e . PBBD `PIID = WL + rK + POO + FT (16) where d = O s is, as in (6) . a function of FT. Equation (16) can also be rewritten as PBBs - Pi (Is + BI) = wL + rK + Po(13D + Xe) + NF (i6)° The model of the North consists of the same 15 equations, but with possi- bly different parameters a. Q . al, a2, b 1 . b2, c 1 , c2. The following equation now substitutes for the original eqn. (6) : and . of course, the equations corresponding to (8) and (9) reflect the fact that the North imports oil and exports industrial goods . In a world trade equili- brium the prices of the traded goods must be equal : Pe = Pa (17) P1 = Pr (18) and traded quantities must also match: XV = 11 a (19) Xf = B? (20) where Xr and Xf represent. respectively, the North's exports of I and imports of oil. There are therefore two sets of eight exogenous parameters each, one set for the North and the other for the South. Each set contains a, Q, 9 1 . a2. b . 1 . b 2. c l , and C2. These parameters are generally different is the two regions. We shall make certain stylized assumptions to simplify computa- tions : a is large in the South and relatively smaller in the North, indicating that labor is more "abundant" in the South. The corresponding parameter for capi- tal exhibits the opposite behavior: P is larger in the North than in the South. We shall also assume that c 1 is small in the South i.e. the production of basic goods uses little capital . and a2 is small in the North. i.e. Northern industry uses little labor . There are a total of 33 independent equations for the com- plete North-South system : thirty correspond to two sets of (1) through (15). one set for each region, and three equations arise from the international trade conditions (1T) through (20), since of these four. as usual, only three are linearly independent . There are 17 endogenously determined variables each in the North and in the South: p1-P ,# , PB- ti, r, Ls . LD . KS . KD . BS . BD . IS , ID- M?-'OS , 'OD . and XS . Finally, we have the transfer FT, making a total of 35 endogenous variables for the complete North-South system . We therefore have 33 equations in 35 unknowns . When we choose the aurneraire (p,#'= 1) an equilibrium is determined up to one variable . If we fix exogenously one vari- able, the equilibrium is (locally) unique . We choose this variable to be the value of the transfer FT. The transfer or loan thus becomes a policy variable . In the Appendix we show how to compute explicitly a solution to the model, i.e . a value for each of the endogenous variables . for each policy sector FT. In par- ticular, we show that by successive substitutions the more important proper- ties of the model can be obtained from the study of a single equation, giving an implicit relationship between the financial transfer FT and the price of indus- trial goods relative to oil. There are a number of determinants whose signs are important in the fol- lowing sections, which determine factor intensities in the different sectors . In total we have the following technical input-output coefficients: a1 b1 c1 a2 b2 c2 in each t'egion . The determinants to be used are : D = aIc2 _ a2c1 N = clb2 _ bIc2 Q = a2b1-a1b2 The assumptions are : DN > 0. DS > 0. NS < 0, QM < 0 The positivity of the determinant D implies that the basic goods sector is rela- tively more labor intensive and the industrial goods sector relatively more cap- ital intensive . The assumption (made above) that the basic goods sector uses very little capital in the South implies that cs is small and therefore that XS < 0 . The industrial goods sector in the North was assumed to use little labor : hence a2 is small and Q1v < 0. The above assumptions on the signs of  -In- the various determinants are maintained at all points below unless there is an explicit statement to the contrary . 4. LAIN RESULTS : TRADE AND DEBT This section studies the impact of a change in the net transfer FT on the economies of the North and the South. Before going on to the algebra, it seems useful to explain the economics of this impact. An increase in the transfer FT increases oil supplies %3S. since the South invests borrowed funds in expanding the oil sector. At the new equilibrium . corresponding to higher FT, the total amount of oil utilized in the North and in the South therefore increases . This in turn alters the supplies of both goods in each region, possibly in different proportions . The composition of the product changes in both regions. The changes in supplies lead to new equilibrium prices for the two goods . The prices of the factors labor and capital also change as relatively more or less labor and capital are employed. This implies that total income in the North and in the South are different at the new equilibrium . The results in this section give simple sufficient conditions for determining the signs of each of these effects . The first theorem gives conditions under which an increase in oil supplies decreases the price of oil with respect to that of the industrial good. While it is intuitively plausible that the price of oil should drop as supplies increase, this is not always true . The second theorem gives conditions under which the rela- tive price of oil increases as the transfer increases oil supplies . Whether one or the other result obtains, depends on the relative strength of supply and demand effects, and the general equilibrium solutions trace this in detail. The results are obtained from various assumptions on technologies and initial prices. The next step is to explore the general equilibrium impacts of an increase in the relative price of industrial goods . The rate of profit rises both in the North and in the South . In the North. the rate of profit and the real wage move together, because the North's economy is rather homogeneous . Therefore, both wage and profit income increase in the North . and we show that there is also an increase in the consumption of both goods, even allowing for the loss of national income due to the transfer . All this occurs because the transfer has improved significantly the North's terms of trade . In the South . because of the rather different technologies in the two sec- tors, the real wage moves in the opposite direction to the rate of profit. The transfer increases oil supplies and oil exports, but oil revenues in terms of industrial,,goods imported are reduced. Wage income and domestic consump- tion of basics decrease as well. If one sought to improve wage income without negatively affecting industrial consumption in the South, the economy of the South would have to be made more homogeneous . The second theorem explores a different set of assumptions, and arrives at rather different conclusions . Now the transfer increases oil supplies, but it also increases the relative price of oil with respect to industrial goods As the terms of trade of the South improve, its macro variables react differently, and so do the variables in the North . The conditions under which one or the other result obtains are therefore quite relevant for policy, and should be deter- mined empirically . The simulations in the next section are a first move in this direction.  A factor that plays an important role in determining the results of an increase in the transfer FT is the sign of the expression A= (c 2 /D-2w/p s a ) where D is the determinant of the matrix The role and interpretation of this term have been discussed elsewhere (Chi- chilnisky 1981a.b) . Basically, the sign of this expression determines whether income effects will dominate price effects, so that increases in supplies will be proportionately larger or smaller than increases in demand as prices change . We refer to an economy as dual if c2/ D < 2w / pg , since a large D would have this interpretation . Conversely, the economy is homogeneous if c 2/ D > 2w /pq . It should be noted that this condition can be written so as to be independent of the particular units of measurement used . Theorem 1 . Consider a North-South economy as defined above . Assume the economy of the North to be homogeneous (c2/ D > 2w /pB) and that of the South to be dual (e2/ D < 2w /pB) . Suppose that at the initial equilibrium the price of industrial goods and the rate of profit are relatively high in the North (p1 > b 2 and 2r > a I / D) . Labor is relatively abundant in the South (a large) and capital relatively abundant in the North (ft large) . In this case an increase in the transfer FT to the South has the following consequences : (i) Oil supplies and oil exports increase in the South . (ii) The North exports. and the South imports. fewer industrial goods . However . the terms of trade move in favor of the North (pt increases) so much that its export revenues rise . There is a corresponding fall in oil export revenues of the South denominated to terms of its import I . (iii) Profits and real wages rise in the North, so much that its consump- tion of both goods increases. (iv) /n the South . profits rise . but employment, real wages, and consump- tion of basics all fall . Proof. We consider first the market-clearing condition in the oil market: XS = .0 (21) From (6). (9) . and (6)' . this equals : -ii(FT) - di = 3§ (22) From (14), = b '3D jBS T b21S (23) and from inverting ( :2) and (13) we obtain : 'OD = (c2L - a2K) + ~(a1K-c1L) (24) In view of (3) and (4) . we may rewrite (22): N N N .OS(FT) - _r Q E- (25) + n Pe + D nN (P where L and Q are the determinants defined above. Equation (25) gives an implicit relation between real wages and the rates of profits in both regions . and the transfer FT . which we denote as / 9P [r N . r 5. (ti pB)N. (W I pg) S . FT) = 0 (26) Since factor prices are functions of commodity prices (see Appendix eqn A .7), we obtain from substitution of (A .7) into (25) a function linking the transfer FT to the prices of B and I : M ~p , ,6S(FT) + ' c 2P - c 1PI + M) + 2(PlaI -Plat + Q) D (27) pa % N N + N (D~ )2P czP' - c A, r + MN) N + (D ) 2 Q(P1aN®PBo2 + Q) = 0 Equation (27) is an implicit function of the form r(FT .p1 , p8"P9) = 0 However, the prices of basics pB and PB (which may be different since basics are not traded) are themselves functions of the price of industrial goods p1 in equilibrium From the Appendix eqn (A .13) we obtain : PB = PB(PI) and PB = PRPI) Therefore, eqn. (27) is actually an implicit function of p l and FT only r(FT.pl) = r(FT.pl . pi(PI), Pff(PI)) = 0 (28) It is then possible to differentiate implicitly across equilibria and obtain 8pl,8FT . or equivalently its reciprocal I OFT = - Or , OP I OPd 87T1 Or (29) This equat"lon represents the change in the price of industrial goods that fol- lows an increase in the transfer FT . By (27) and (6) . 8r __ 8,3S > 0 OFT OFT Therefore the sign of (29) is always that of -8r/, 8p1  - 13- We may now compute the derivative -8^/ 8pI . From, (27) and (28) we obtain _8P MmM a N N,,Iv ~ )- N-cNPI)_ ~1BV)2 -(DN)2 (DN) a Nc N IV PN Q Na N (DN)2P# (DN)2 From expression (30) we may compute the changes in pl as FT changes . pro- f vided we know the signs of the derivatives Bpi/ 8p, and Op / 8pI across equili- bria . The next step is therefore to compute the signs of the derivatives of the price of .basic goods with respect to the price of industrial goods across equili- bria in each region . For this we utilize the expression relating the real wage and the rate of profit in each region, derived from the market clearing condi- tion Bs - RD = 0: act w fla2r a ( !~) 2 = 0 (31) D PB D PB (see Appendix eqn. (A .11)). and also the equations relating factor prices to commodity prices : = PIaj _PBa2 y B r 32 D () w _ PBc2_PICI + M (33) PB DP B (see Appendix eqn (A 7)) . Equation (31) is an implicit expression between real wages and profits in each region, denoted A(w /PB,r) = 0. Since eqns . (31) and (32) give real wages and profits as functions of commodity prices . (31) actually gives an implicit relation between commodity prices in each region . denoted *(PB " Pl) =AI - -(pB.pj) .r(PY .Pr)I =0 (34) PB From (34) . by the implicit function theorem . in each region- OPB (35) OPI _- ( 8P ,/ ~ iqj~ . 8(u~ lPB) _a'k . 8r ay . 8( u; / PB) 6-~ 8r _ + 8pI 8(w/PB) + br 8pI l a(w/PB) . Spa 8r &PB  ®14° Furthermore, from (32) and (33) we find that the partial derivatives P(w APB) __ <0 (36) 8P1 DPB ar = aj >0 (37) apl D 8(W/") - plc' -~ 0 > 0 when pl > 62 (36) 8PB Dpj and _ 0 (39) 8PB D Therefore we obtain from (35) and (39) 8PB _ ~ c i a jPa2 (PIC T a? + / e - ) + (40) 8P1 DPB D2 , DP B DZ where A = a(c 2 / D -2w IPB) From relation (40) we may now determine the sign of ap B / aPl in both the North and the South. First note that ap B / apl is always positive in the North since p t > b 2, so that p1ci-M > 0, and G > 0 by assumption In the South A < 0, but fl is rather small . Therefore, (40) is also positive in the South. With this information we may now return to eqn (27) and compute pg % apl . As a is large in the South and P is large in the North, we have from eqn . (30) that the expression for -pg i apl is dominated by the following terms : ac ~~! apE a2 PNQX PNQNaN aM 8P B (M -C'PI) (41) ( D N)2 (D N)z D2P 8P1 + DZPB + 8P1 Here M-c1PI = c1 6 2 - Lice - CIPI is negative as c I is small in the South. Hence the first term is positive (because MS < 0) and dominates the second, which is multiplied by c i . As Q'?~ < 0, the third term is negative and the fourth positive But a2 is small in the North. so that the fourth term dominates . Hence we have that - ~U-> 0 aPt This implies that the price of industrial goods pl rises as the transfer to the South increases, i .e OPI (42) aFT > 0 We next study the movements of the rate of return in the North r ^l as pl changes . From the national income identity P11E = rK -FT  5- 1 As lff = lk-Xf and p,Xf = X' = Off. Plik = rX + Og - FT In the North. P is large . We can therefore neglect terms other than those in ~e giving _. P/ = [(-Q /D) + r, l (ai/D) with 'OP! =D/a, > 0 (43) Hence as FT rises, p, rises and the profit rate-in the North rN rises. Knowing how rN moves enables us to find the sign of the change in the real wage in the North. We can rewrite the market-clearing condition for the B market . eqn . (7), as aczw Pa2r_ a _tu (p8I'= 0 DPB D (see Appendix eqn . (A .11)) . Implicit differentiation gives : 8(w / pB) - azB (44) 8r - D6 where t3 = a(c2/ D -2w lpB) . As A < 0 in the North by assumption. we have that d(W /PB ) in the North Hence an increase in FT raises the real wage in the North, as well as the profit rate . The next step is to show that the consumption levels of B and I rise in the North. Ip - rK-FT = Pre-FT 8Iff - 8r N 2dr -1 (46) 8FT 8FT which is positive for large P Also, B1~= wL / pB = a(w / pB) 2 (47) so that B,1qV also rises with FT by (45) . (42) . and (43) We have now proven point (iii) of Theorem 1 . Next we study the response of trade patterns to FT We have, by inverting (12) and (13) . a K c X~= IS - Ip = D - ~L rK + FT From '3) and (4) _ CL I c1w r _ -prZ FT XN _ DPB +  - 1 6- Hence 8Xr 8(w I PB) apt _P(!~L-2r)-cl + 8FT 8r D D 8r apt 8r By the conditions of the theorem, the first term is negative . By (45) the second term is negative. and by (44) it contains P. As Q is large, these terms dominate, and ax,N <0 8r i .e the North's exports of the industrial good fall as FT and hence rN rise . This implies, of course . that the South's imports of industrial goods fall, a MS <0 (49) Sr N We next check what hapens to the volume of oil traded . This equals oil demanded to the North, 131, which from Appendix eqn (A.3) is - awm _ PrQ DP B D Here y4 is large and Q is negative, by assumption r rises, by (43) Hence 813Y _ aX,so (50) -WT 8FT > 0 This proves points (i) and (ii) of Theorem 1 . What remains is to study the behavior of the Southern economy . We first show that rs rises with FT . This is done by showing that BMf / 8r s < 0. As 8Mj _ 8mj Or N _ 8r s Or N Or s this will imply from (49) that 8r N / Brs > 0, which in conjunction with (42) and (43) gives the desired result . Mh -- !~ -13F = rK + ,6s + FT -Is v Pray claw =pr Z - + + -6S + FT D DPB a MS 8(w /P B) 8,3s 8 FT LM-Is ~(2r -a~/ D) + (c l a/ D) ( 8r ) + ( eMjs + 8ddjs ) 8rs Ors - 81dj ( 1- 8Js - 8FT ) = P(2r _ a l / D) + c ~a/ 17 ( a (w P B) ) 8rs amj ams 8 Now "s _ a$s Or N <0 8MI 8r N 8M!  - 17- by (49). (42) . (43), and (S) . Similarly . OFT / 8Al< < 0. By (44), O(w / PB) / Or < 0 in the South. As by assumption as is large. this establishes that < 0 so that Ors (51) Drs 8FT > 0 It now follows from (44) and the tact that AS < 0 by assumption, that real wages in the South fall with FT. It follows immediately from (3) and (5) that employ- ment and the consumption of basics also fall. This completes the proof of Theorem 1 . Theorem 2. Suppose MS > 0, i.e . c lbz-b lc2 > 0 in the South. Let pg be small and Pt > b? at the initial equilibrium, with all other conditions as in Theorem 1 . Then an increase in the financial transfer to the South has the opposite effects to those established in Theorem 1 : it leads to a fall in pt, the price of the industrial good, and a relative increase in the price of oil. even though oil supplies have increased. The oil exporter's terms of trade therefore improve . In addition., oil exports and the rate of profit in the South decrease . The North exports more industrial goods . Real wages, employment, and con- sumption of basics increase in the South. In the North, the rate of profit and the real wage decrease . Proof . As in the proof of Theorem 1 . the sign of aFT / ap t equals that of ar / Op t.This is given in eqn. (30). or approximately in (41). The latter may also be written as am + PQ apg ,a 2 -a N aPB (R - c1Pt) 1+c~ M )2 aPt ~Pa apt PB (D Now note from (40) that for large PN . Op B _ _a i Opt a2 Hence the second term above is zero and (41) can be expressed as OPB 1 OPB b I am ~c~ (b2 Pt) 2 (41) 8pt PO + 1 OPt PB D?pg Under the conditions of Theorem 2. this is negative, proving that the oil exporter's terms of trade improve . i.e. Pt falls with FT. The rest of the theorem follows immediately . Inequality (43) implies that the profit rate in the North falls . and (44) implies that real wages in the North fall. Inequality (48) tells us that the North's exports (and the South's imports) of industrial goods will increase, and from (50) we then know that oil exports of the South fall (52) establishes that the rate of profit in the South falls, and using (44) again proves that real wages, employment, and consumption of basic goods rise in the South This completes the proof . The main difference in the conditions of Theorems I and 2. which reverse the results, are first, the sign of Ms and second. the impact that the transfer has on the relative price of industrial goods. The sign of XS is positive in Theorem 2, and negative in Theorem 1 . It seems more plausible that YS should be negative, since this happens when the basic goods sector in the  - 1 8- South uses few capital inputs . Theorem 2 assumes, instead, that the basic goods sector is more capital intensive. The impact of the transfer on prices seems also more plausible in Theorem 1 . There, the transfer increases oil sup- plies, and this leads to lower oil prices. In Theorem 2. the transfer also increases oil supplies, but this leads to higher oil prices. Clearly, an empirical analysis of the actual conditions is needed to evaluate the results, but. a priors, the conditions in Theorem 1 appear more intuitively natural than those in Theorem 2. A final point is the stability of the equllibria under the standard Walrasian adjustment process in which prices increase with excess demand, and decrease with excess supply . This is a rather specialized issue since the model has con stant returns to scale . The Walrasian stability of a closely related model (Chi- chilnisky 1981b) has been studied in Heal and McLeod (1983) and the interested reader is referred to that paper for a detailed analysis. 4 . CONCLUSIONS We have considered a situation where an inflow of capital investment into a country's oil sector has allowed that country to run a deficit on its balance of trade The capital inflow is, of course . matched by an accumulation of indebt edness to foreigners . An inflow of foreign capital . whether used for consump- tion or for investment, inevitably affects the internal equilibrium of the receiv- ing country . Consumption patterns, production patterns, and prices all change. The same is true of the lending country it changes its consumption pattern by making a loan. and for this reason, and because the equilibrium of its trading partner changes, its own domestic equilibrium alters . A crucial fac- tor in determuxing these macro effects of a loan is the change in relative prices (oil prices, industrial prices, and prices of basic goods that are not traded) . A loan must be of a significant size before having a measurable impact on prices, and the cases we discussed here, where the loan is of the order of 100 billion US dollars, certainly fit this description. It is clear, then . that it is a complex matter to trace the full impacts of a loan from one trading country to another. Our model has enabled us to iden- tify these impacts in a rather simple fashion, because of our somewhat stylized assumptions, and to assess the gains and the losses arising from such a loan for different groups within the lending and borrowing countries . One important feature to emerge is that the loan may have a beneficial effect on the equili- brium of the lending country. This happens when the borrowed funds are used to increase oil supplies, leading to more abundant oil. increased oil exports, and lower oil prices . The terms of trade of the lending country improve, and this leadsto higher levels of consumption of both goods in the lending country . Theorem 1 establishes the conditions under which the welfare level in the lend- ing country will rise as a result. In making a social cost-benefit analysis of such a loan, this is a point that should clearly be considered: there is a social return to the loan over and above the rate of interest paid on it . It is possible that even if a major rescheduling that delayed repayment were to happen, the lending country as a whole could nevertheless benefit . Private financial institu- tions making the loan might of course be strained in such a situation. There could then be an argument in favor of the government compensating banks in the case of temporary losses. in view of the positive externalities that their actions have generated for the rest of the economy . Obviously, such a policy would require very careful analysis of the macro effects and of the interna- tional markets concerned.  _19- Similar issues apply to the receiving country . The borrowing sector may benefit in commercial terms from the loan, but a social cost-benefit analysis of the loan should also take into account its effects on the overall economic equilibrium . As Theorem 1 shows, these could be substantially negative . if there has been overspecialization in one sector thus leading to lower terms of trade for the country, with correspondingly negative welfare effects . In sum- mary, the fact that a loan, if large, may affect the equilibrium pattern of prices and quantities in both countries means that it will have macroeconomic conse- quences going tar beyond its impacts on the profits of the borrowing and lead- ing institutions . Theorems 1 and 2 have indicated two very different possible outcomes . In one case, the effects are beneficial to the lending and harmful to the borrowing country, while in the other case the opposite is true . The disUnguishuig feature is the effect of the loan on the oil exporter's terms of trade. In the first case, they worsen, and in the second, they improve . Which of these two outcomes occurs depends on the patterns of factor intensities in the receiving country and the initial price levels . Once these are known, thus establishing whether the loan improves or worsens the receiver's terms of trade, everything else can be traced . Experience indicates that over the last three years, the terms of trade of oil exporters have worsened. While many factors have contri- buted to this price movement, this suggests that a policy of borrowing to invest in the oil sector might not have been the most favorable to the oil exporter . However, such a policy could be favorable to the lender ; it yields more oil at lower prices. Such macro outcomes should be computed when discussing the present situation. The calculus of the debt must go beyond the financial aspects, and must include the macroeconomic effects on prices, imports, and exports of both countries . It is important to emphasize that we have studied the consequences of granting a loan before thus was repaid The repayments will not have effects that are simply equal and opposite to those of the granting of the loan. The asymmetry arises because . when the loan is made, it is invested or consumed in sectors different than those that will pay the debt . For instance, in this paper the debt was used to build up the production capacity of the oil sector However, when the loan is repaid, this will not of course coincide with running down this capacity . Investment is irreversible, and capital stock and machines depreciate . The loan will be repaid by running a balance-of-trade surplus. The effects of running a trade surplus at a constant capacity level in the oil sector are not the opposite of those running a trade deficit and using the capital inflow to expand oil-producing capacity . As a matter of fact, both could affect the major macro variables in the same direction . This distinction between receiving and repaying a loan will be developed further in a subsequent paper . Finally . we point out a connection between the problem that we have stu- died here and the extensive literature on the transfer problem in international economics. This literature is concerned with the possibility that a transfer of resources from one agent, or country to another may benefit the donor and harm the recipient. This issue has so far been studied only in the context of a barter economy without production in the case of perfectly competitive gen- eral equilibrium models For surveys of these results, see Chuchilnisky (1980) . Jones (1983) . and Geanakoplos and Heal (1983) . Our present Theorem I pro- vides an example of the transfer paradox in a production economy resources are transferred from lender to borrower, and the lender gains as a result (Theorem 1). even though the receiver expands its production capacity .  20 APPENDIX 1 : Analytic Solutions This appendix gives an explicit analytic solution to the model . and presents the results of numerical simulations on the effects of rescheduling the debt reported in the paper. In order to solve the model we consider first the equation equating oil exported with oll imported: XS - Me In view of (6) . (9) . and (6)'. this equals 3 i3s(FT) - 'OD = ' D (A.2) where the left-hand-side variables are from the South From (14). (12), and (13) = -'(c LL -a2K) + ~a1K-cIL) (A.3) 'dD D D PS D where M =cjb2-b1C2 Q= a2bI-alb2 Therefore, we may rewrite (A.2) as ,3S(FT) - a '-M - Er Q = °` N ( w )NMN - n'rNQn' (A 4) D PB D DN PB DN (A.4) is therefore an implicit equation in five variables, which we denote (A .s) Our next step is to write the rate of profit r and the wage w lpB in the two regions as functions of the prices of basic and industrial goods . PH and p, . Recall that oil is the numeraire (p d = 1) . From the production functions (1) and (2) we obtain the associated competitive price equations pa = a l'c' + b ipe + c'r (A. 6) PI = a2w +b2Pe + c2r (pB-b2I = ~a2 ~2 since p,6 = 1 . We therefore obtain the factor-commodity price relations :  - 2 1- +M __ c2PB_CIPl "' D 1)c2 + (b 2 -P!)c I _ _!!L_ __ (PB b (A.7) PB DPB _ __ (bI -PB)a2+ CLIV"I - b2) __ CLIP/ CL2PB + Q r D D Substituting w / pB and r from (A .7) into (A5), we obtain a new Implicit func- tion. in four rather than five variables : *(FT. pl . PB. PB) = 0 (AB) Recall that pa may be different from pB because B is not traded internation- ally . The last step is to substitute pB and pB as functions of pl into (A .B) . This will lead to an implicit function in two variables X(FT . Pl) = 0 (A9) Since FT is an exogenously given parameter, (A.9) is an analytic solution to the model . from (A.9) we may compute the equilibrium level of industrial prices pj(FT) . 1t is easy to check that once pJ is known, we may solve for the equili- brium values of all other endogenous variables . This will be explained below Now, in order to obtain pB =PB(p/), we use another market-clearing con- dition . this time in the B-market : BS = BD (A.10) From (12) and (13) this can be written as c 2L - a2K = L D PB or a ct w _Pazr-a(- 2=0 D PB D PB from which we obtain _ W c2 c2 Pa 2r 2 (A.12) FY B D _14D2_ Da ) _ a two-branched function relating w / P B* and r The different parameter values will determine which is the appropriate branch in (A .12) . Usmg again the factor-commodity price relations, (A.12) yields an implicit relation between pB and p/, as desired : 1 2 c2 1 c 2 _ Pa2 _ 2 _ c2 + c 1P/ c (Q 2pB + IPl) M 0 aDPB t D PB 4D 2 2a a DPB D DPB (A.13) Substituting (A.13) into (A.8). we obtain the desired relation (A.9) between FT and pt y(FT . pt) = 0 From (A.9) we may then compute p! =pj(FT) . From (A.13) we obtain ps(N) and p;(S) . and from these three equilibrium prices we obtain the equilibrium rates of profit r"(N) and r "(S) . and of real wages. (w / pB) "(N) and (w/pB) " (S) . From these we obtain supply of labor and capital in the North and the South. and using the inversion of (12) and (13) we obtain the output of B and 1 in both regions . From the national income identity we may compute demand for 1 in the South, which determines imports from the North . and from (40), exports of oil from the South . From (14) we obtain oil demanded in the South, thus completing the computation of the equilibrium .  - 2 3- APPENDIX 2 : A Simulation for Mexico and the U .S .A . 2a . Listing of Equations of the Model :System of simultaneous equations solved by TK-solver software package 2b . Listing of Exogenous parameters and of endogenous solutions Numerical values of both 2c . Background Data for the Model 2d . Results of Simulations  - 24- A . Listing of Equations of the Model : System of simultaneous equations solved by TK-solver software package " U .S .-Mexico Model with F\nanicial Transfers * D=a1*c2-c1*a2 "Misc . Determi nants * Ds=a1s*c2s-c1s*a2s * M=c1*b2-c2*b1 * Ms=c1s*b2s-c2s*b1s * Q=a2*b1-a1*b2 * Qs=a2s*b1s-a1s*b2s * w=(M+c2*PB-c1*Pl)/D * ws=(Ms+c2s*PBs-c1s*Pl)/Ds * r=(a1*PI-a2*PB+Q)/D * rs=(a1s*PI-a2s*PBs+Qs)/Ds * K=b*r+SK "Capital Supply * Ks=bs*rs+SKs * L=a*(w/PB)+SL "Labor Supply Equations * Ls=as*(ws/PBs)+SLs * BS=(c2*L-a2*K)/D * BSs=(c2s*Ls-a2s*Ks)/Ds * lS=(a1*K-c1*L)/D * ISs=(a1s*Ks-c1s*Ls)/Ds * NT=Fn-Fs "Net Financial Transfer * OSs=(NT*sh)*k+SOS "Oil supply depends on NT and k (capital output * OS+OX=b1*BS+b2*IS "oil supply = demand * Bd=(w*L)/PB " Basic Good, S=D * Bds=(ws*Ls)/PBs " Basic Good, S=D * BS=Bd * BSs=Bds * lds=lSs+Xl " Industrial Good, S=D * Id=IS-XI * L=LD "Equilibrium Conditions * K=KD * Ks=KDs * Ls=LDs * Pl*Xl=PO*OX "Balance of Trade * TOS=OS+OSs * TOD=b1*BS+b2*IS+b1s*8Ss+b2s*lSs * TOS=TOD * TIS=IS+lSs * lId=ld+lds * wexp=PO*OX-Pl*XI * PBs*BSs+Pl*(lSs+Xl)=ws*Ls+rs*Ks+PO*OSs "Walras , Law * PB*BS+PI*(I8-XI)=W'*L+r*K+PO*OS * Sdual=c2s/Ds-2*ws/PBs "Duality Indicies * Ndual=c2/D-2*w/PB * NGNP=PB*BS+Pl*lS * SGNP=PBs*BSs+PI*lSs * PP=1/Pl 2b , Exogenous Parameters and Endogenous solutions  ®26- St Input Name Output Unit Comment U .S . Mexico Test Data 1:4-84? ,5 .55 SOS Mex oil supply intercept L 53 .7875 Fn $80 Financial flows from the U .S . 5 Fs Financial flows from Mexico 1 ..86 1 k Petroleum capital output ratio 1 =_.h Share of capital inflow invested in pe 9 .0484549 MBD U .S . oft supply 1142 .4624 'LK U .S . Capital Intercept 106 .32042 SL U .S . Labor Intercept 8 .052052 a U .S . Alpha 3783 .0258 b U .S . Beta 13 .388079 as Mex Alpha 955 .61479 bs Mex Beta 11 .54 SLs Mex Labor Intercept 246 .7{281 SKs Mex Capital Intercept .08179253 a1 U .S . labor input .06475242 a2 U .S . labor input .0649''331 bl U .S . Oil input .02782428 b2 U .S . . Oil input .74613511 ci U .S . Capital Input 1 .3952113 c2 U .S . Capital Input .32017112 als Mex Labor input .15762271 a2s Me>. labor input .1899686 b1 s Mex OIL input .06449843 b2s Mex OIL input 1 .2904133 c1s Mex Capital input 3 .5938466 c2s Mex Capital Input L Id 626 .23747 Capital Goods Demand in the U .S . L PB 1 .1730468 Price of basics in the U .S . L PBs 2 .0886154 Price of basics in Mexico L PI 1 .1261535 World Price of Capital Good=_. 32 PO $80 Price of petroleum L PP 28 .415310 $80 Terms of trade L OSs 3 .7457175 MBD Mexico's Oil Production L Ox 2 .0213063 MBD Oil Exports L 20 .964166 Capital Goods Exports L ws 5 .9885036 Mex wage rate L rs .03275902 Mex Profit Rate L K 2181 .9551 U .S . Capital Good=_. Supply L L 182 .11085 U .S . labor supply L r .27477811 U .S . Profit Rate L w 11 .041374 U .S . wage rate L BS 1714 .1293 U .S . supply of Basics L Bss 143 .14976 Mex Basic Good Supply L is 647 .20163 U .S . Capital Goods Supply L Iss 25 .973689 Mex Capital Goods Supply L Ls v 49 .926463 Mex Labor Supply L k:s 278 .06781 Mex capital stock Bd 1714 .1293 U .S . Basic Goods Demand Bds 143 .14976 Mex Basic Goods Demand Ids 46 .937855 Mex Capital Goods Demand LD 182 .11085 U .S . Labor Demand LDs 49 .92646'3 Mex Labor demand KD 21 ;31 .9551 U .S . Capital Demand KDs 278 .06781 Mex Capital demand NT 43 .03 $80 Net financial transfers M - .0698211 Ms - .3803653 D .06580381 Ds .94724745 a .00192812 Os -3 .177E-4 2c' . Background Data  "VIIEXICAN AIL EXPORT PRICE SOURCE: WHARTON ECONOMETRICS $40 .00 $35.00 -i $30-00 - W $25.00 $10-00- $5-00 -! $0 .00 70 ~2 74 7e 78 so 82 84 YEAR  -29® NET CAPITAL INFLOW AND PEMEX INVE TME SOURCE: WHARTON ECONOMETRICS B 5 4 O 3 0 71 73 75 77 79 81 83 YEAR 0 NET CAPITAL INFLOW + PEMEX INVESTMENT MEXICO OIL PRODUCTION AND EXPORTS SOURCE: WHARTON ECONOMETRICS 3.5 2-5- 0.5 -~ 0 70 72 74 7E 78 80 82 84 YEAR 0 OIL EXPORTS 4 OIL PRODUCTION 2d . Results of Simulations  ®31- i'w :aOS9 :bi > b2') Northern Variables i= n pp FS Pes PI I ci Ws 32 .2725 34 .0888178 .914516841 " .7 14766882 .938724252 820 .064008 1 .03146752 3:3 . ;.14825 33 .75740 70 .927438634 .733417546 .947940108 E: i 4 .1 ::.5785 i .08'167339 34 .424 33 .4331172 .940307566 .7541 37708 .957134801 . .68899 80i l .15575974 35 .49'975 33 .1186351 .953005310 .77745736i .966223394 790 .259246 i . G.3 i 85307 36 .5755 32 .81 78878 .965354568 .804166411 .975077988 779 .144550 1 .321i2451 '37 . 65125 32 . 537003.2 .97707470i . 8;:55330 i i . 93'3495524 768 .838783 1 .42863204 :38 .727 :32 .2865524 .987680855 .873825090 .99112471i4 759/ . 7 i O987 1 .56354870 39 .80275 :32 .0 878052 . .996200128 .923862717 . 5'97260 T98 752 .5i2537 1 .74503187 40 .8785 31 .999'1827 1 .00000077 .999'115617 1 .00000054 749 .3387'14 2 .02384347 41 .5.5425 32 .3708998 .98405'23;34 1 .20495038 .988542i8c. 762 .7785i4 2 .82335258 4;3 .03 32 .5135153 .5'78063190 1 .42562704 .984206090 767 .980182 3 .66650443 44 .10575 32 .2540031 .989070197 1 .53002495 .992124914 758 .5259014 4 .05407772 45 .1815 31 .9148182 1 .0037007 1 .61056996 1 .00266904 746 .271 1 '16 4 .64302423 46 .25725 31 .5,377292 1 .0203005c/ 1 .67965860 1 .01465771 7 ;.12 .771003 4 .53331 :390 47 .333 31 .1362644 1 .03837379 1 .74292248 1 .02774050 718 .546613 4 .81101399 48 .40875 ;30 .7164441 1 .05773147 1 .80281566 1 .04178726 703 . 8:37644 5 .01911250 49 .4845 30 .2814013 1 .07830509 1 .86076714 1 .05675427 688 .774243 5 .218590 57 50 .56025 29 .83291 24 1 .10008564 1 .91771559 1 .07264033 673 .436826 5 .41301048 51 .636 29 .3720246 1 .12309931 1 .97433732 1 .08947191 657 .8713418 5 .60490413 52 .7 1175 23 . 8'"193751 1 1 .147399651 2 .03115332 1 .10729038 1,42 . i 36200 5 .79620309 5 ;3 .7875 28 .4153098 1 .17304680 2 .081301544 1 .1261534'3 626 .237469 5 .98850301 NSMOS10 List of Variables rs r w OX OSs Sdu a 1 SGHP .193017110 .296207419 7 .03508732 1 .07975067 2 .15325243. .9078243.82 130 .702837 .198028599 .294947169 7 .45450074 1 .135`77608 2 .23287571 .822493933 1 :3;3 . :358107 .197688538 .29'3712032 4 .02 :316481 1 .19786324 2 .31249896 .728873305 136 . 1'18653 .196440119 .2925146013 8 .1493 ;3631 1 .25:361800 2 .39212222 .6250 615 :38 139 .263552 . 1 ?54285799 .291368766 4 . :.=:5077224 1 .30640784 2 .47174547 .508289901 142 .636193 .193054292 .290 25'3685 4 .50342435 1 . ;35524371 2 .55136872 . :374227100 146 .4i04i4 . 1 89201 1,39 .249344389 8 .04219709 1 . ;:9340979 2 .03099197 .215:.158381 150 .324024 .183008845 .244588736 8 .75 :32520 :3 1 .43239492 2 .71061522 .0163020'70 156 . ;36107,3 .171279621 .284254251 8 .80277019 1 .447 ;36302 2 .79023444 - .2060403 i64 .576'30i .1 :33288736 .245'030055 4 .595:39340 1 .34 :391228 2 .80980173 - .89220010 188 .373944 .095102303 .290209204 8 .51072049 1 .35930759 2 .5'4948498 - i .34'171 85 7 2P-' .241237 .040307203 .289220969 8 .66031370 1 .40399350 3 .02910423 -1 .50 :32951 235 .503761 .070504220 .247930314 8 .45090075 1 .46142149 3 .10873148 -1 .5991627 244 . -31 L:609 .063124006 .246497241 9 .06694443 1 .52534407 3 .18835474 -1 .0099944 259 .604304 .057014452 .2449744'" ;5 9 .30184029 1 .59210292 3 .20797799 -1 .7" 266396 270 .032809 .051800406 .243385824 9 .55300007 1 .66094877 3 .34760124 -1 .7740925 280 .044441 .047210108 .281744553 9 .41950641 1 .73126400 3 .4272449 -1 .8150451 28'1 . !5727 i .043109525 .240054012 l0 .1011763 1 .80267020 3 .50684774 -1 .3512401 299 .340518 .039376555 .278;33 -3200 10 .3982818 1 .87491925 3 .58647100 -1 .8837682 08 .850221 .0 :35944108 .2765722J0 10 .7114057 1 .94783931 3,00009425 -1 .91 :33040 318 .505083 .03,275901 7 .274774114 ii .0413738 2 .02180624 ;3 .7457175 -1 .9404351 ;.123 .235104  aital Transfers and h~~~e ; ican °N!~ U.S .-Mexico Test Data 500-0 450-0-4 400.0 350.0 300.0 250 .0 200.0 150.0 100.0 50 .0 31 33 35 37 39 41 43 45 47 49 51 53 Net Capital Transfers $1980 Billions Capita( Transfers and Oil Prices U.S .-Mexico Test Data $40.00 $39.00 $3$.00 $37.00 $3e.00 $35.00 _ $34.00 $33.00 ®$32_00 I- S31-00 m $30.00 $29 .00 a $28.00 IV $27.00 $28.00 $25.00 $24.00 $23 .00 $22.00 $21 .00 $20.00 31 33 35 37 39 41 43 45 47 49 51 53 Net Capital Transfers $1980 Billions  Capital Flows and U .S . Employrr;ent U.S_-Mexico Test Data 110_0 1 OB.0 108.0 104_0 102 .0 100 .0 0 98_0 w 98_0 0 C 94.0 0 92-0 90 .0 88 .0 813 .0 84 .0 82_0 80_0 21 23 25 27 29 31 33 35 37 39 41 Net Capital Transfers $1980 Billions Capital Flaws and Mexico's Oil Exports U-S.-Mexico Test Data 2-4 2.3 2-2 . .2 -1 2.0 1_9 Q a. 1-8 1 .5 a w 1-4 0 1 .3 0 1-2 1 .0- 0-9- 0.8- 0-7- ,. 0.8- 0.5 31 33 35 37 39 41 43 45 47 49 51 53 Net Capital Transfers $1980 Billions  Basic Prices and Duality U.S .-Mexico Test Data 2.5 2.0 1 .0 0.5 0.0 -0.5 -1 .0 -1 .5 -2.0 31 33 35 37 39 41 43 45 47 49 51 53 Net Capital Transfers $1990 Billions 0 Pb in South 4 c2/0-2w/ab (south) Net Capital Tranefere and U .S . GNP U.S .-Mexico Test Data 3.5 3.0- 2-5 0 M Cf 2.0 -{ 1 .0-4 0.5- 0.0 31 33 35 37 39 41 43 45 47 49 51 53 Net Capital Transfers 51980 Billions