The Valuaion of Hidden Asses in Foeign Tansacions: Why Dak Mae Maes IT IS MORE IMPORTANT THAN IS SOMETIMES THOUGHT By Ricado Hausmann and Fedeico Suzenegge Ricado Hausmann is dieco of Havad's Cene fo Inenaional Developmen and pofesso of he Pacice of Economic Developmen a he Kennedy School of Govenmen. Peviously, he was chief economis of he Ine- Ameican Developmen Bank, Minise of Planning of Venezuela, and a membe of he Boad of he Cenal Bank of Venezuela. He was he chai of he IMF-Wold Bank Developmen Commiee and economics pofesso a he Insiuo de Esudios Supeioes de Adminisacion (IESA) in Caacas. He holds a PhD in economics fom Conell Univesiy. Fedeico Suzenegge is a visiing pofesso of public policy a he Kennedy School of Govenmen of Havad Univesiy and a pofesso a Univesidad Tocuao Di Tella in Agenina. He has held posiions as an assisan pofesso of economics a UCLA, chief econo- mis of Yaciemenos Peolifeos Fiscales of Agenina, dean of he business school a Univesidad Tocuao Di Tella, and seceay of economic policy of he Republic of Agenina. He is a consulan o copoaions and inenaional oganizaions. He holds a Ph.D. in Economics fom MIT. In 2005, he Wold Economic Foum of Davos seleced him as one of is Young Global Leades. This pape claifies how he valuaion of hidden asses wha we call dak mae changes ou assessmen of he U.S. exenal imbalance. Dak mae asses ae defined as he capialized value of he eun pivilege obained by U.S. asses. Because his eun pivilege has been seady ove ecen decades, i is likely o pesis in he fuue o even o incease, as i becomes leveaged by an inceasingly globalized wold. Once his is included in fuue pojecions of U.S. cuen accouns, he U.S. exenal posiion looks much moe balanced han depiced in official saisics. The Valuaion of Hidden Asses in Foeign Tansacions: Why Dak Mae Maes Business Economics Januay 2007 29
I is well known ha he U.S. economy has been unning inceasingly lage cuen accoun deficis since he ealy 1980s. Cuen accoun deficis signal an economy ha is spending beyond is means; so i comes as no supise ha he accumulaion of deficis duing his peiod, adding up o 5.27 illion dollas beween 1982 and 2005, has significanly inceased U.S. ne foeign deb now moe han 20 pecen of GDP. If hose ends wee no in hemselves cause fo concen, in ecen yeas he deficis have escalaed in boh nominal value and as a pecenage of GDP, suggesing ha he pocess canno coninue much longe and ha a lage and painful evesal may be nea. In a seies of aicles (2005, 2006a, 2006b HS hence), we have agued ha his concen may be misguided. Ou wok sas by poining ha such a lage incease in deb needs o be econciled wih a ahe conadicoy fac: ha wha he U.S. economy pays on is ne foeign posiion seems o have been supisingly consan in spie of he measued incease in ne foeign liabiliies. 1 Paaphasing Bill Cline (2005) we asked in ou wok if i made sense o call a couny ha makes money on is ne foeign posiion a debo. The quesion we aised in hese sudies was whehe hee wee hidden asses o sevices (whose size had inceased seadily ove ecen decades) povided by he U.S. economy, explaining why he ne income flow had emained sable in spie of he incease in measued deb. In ha wok we called hese asses dak mae and povided evidence fo hei exisence. In hei aicle, Boowing wihou Deb? Undesanding he U.S. Inenaional Invesmen Posiion (his issue), Mahew Higgins, Thomas Kligaad, and Cedic Tille (HKT hence), acknowledge he exisence of hese inangibles, bu ague ha dak mae is unable o accoun fo he diffeence beween he sock and flow daa, ha he mehodology fo compuing i is doubful, and ha i has no subsanive implicaions fo he fuue evoluion of U.S. imbalances. Naually, we believe hese claims o be incoec. Afe a yea of wok by us and a numbe of ohe scholas, hee is gowing evidence, boh a he mico and maco level on he exisence of significan amoun of dak mae. 2 Fuhemoe, we believe ha hinking in ems of dak mae adically changes he way 1 In Hausmann and Suzenegge (2006b), we show he Unied Saes is an oulie. As fo mos ohe counies, hee is a easonably good link beween cuen accoun imbalances and changes in ne invesmen income. 2 In Hausmann and Suzenegge (2006b), we povide coss-couny evidence on he exisence of dak mae, bu ohe auhos have povided alenaive empiical esimaes. Meissne and Taylo (2006) pesen evidence a he maco-hisoical level. Buie (2006) povides in which we evaluae he fuue evoluion of U.S. and global imbalances. To discuss why we believe ha dak mae is a sound and meaningful concep and heefoe disagee wih HKT is he objecive of his bief pape. Wha is dak mae? Befoe addessing HKT s concens le us fis biefly claify he meaning of dak mae. As menioned above, ou moivaing fac is ha ne income fom foeign asses seems o be pooly accouned by he change in foeign asses obained fom accumulaing he cuen accoun o fom diec measues of he sock of ne foeign asses ha some counies esimae. Thus, we popose an alenaive way of measuing he cuen accoun, one ha sas by defining ne foeign asses of he couny (NFA) as he capialized value of he ne invesmen income (NII), discouned a a consan ae of inees (): NII (1) NFA = The inclusion of dak mae is indicaed by he supescip. Ou use of he concep of dak mae coesponds o ha used in physics o accoun fo he fac ha he univese is moe sable han you would hink if i wee held ogehe only by he gaviy emanaing fom visible mae. In he same way ha physiciss infe mae in he univese fom is gaviaional pull (bu no fom adding up he visible mae), we infe he asses fom hei euns and no fom adding he cuen accoun imbalances. As a esul, counies wih ne invesmen income lage han wha is pesumed on he basis of hei asse base will have dak mae asses, while counies fo which he ne invesmen income is oo low will have dak mae liabiliies. Choosing o value he asses on he basis of hei euns is jus like valuing a company by calculaing is eanings and muliplying by some pice-eanings aio, o valuing a popey based on is enal value. We know fom he copoae finance lieaue ha fo an individual company he eanings of any given yea may give an uneliable measue of is ue eaning poenial, bu if we aveage ove an economy and look a ends ove a couple of yeas, his simple mehodology delives easonable esuls. Of couse, his opens many mehodological ques- an esimae fo liquidiy dak mae. A he mico level Chai and Tesa (2006) povide evidence of dak mae associaed wih fim akeoves in emeging counies. Campbell e al. (2006) povide a jusificaion fo insuance-based dak mae. Theoeical jusificaions fo some ohe channels ae povided in Caballeo e al. (2006) and in Mendoza e al. (2006). 30 Business Economics Januay 2007 The Valuaion of Hidden Asses in Foeign Tansacions: Why Dak Mae Maes
ions bu we efe he eade o Hausmann and Suzenegge (2006a, 2006b) fo fuhe discussion. 3 Wih his measue of ne foeign asses, we define he cuen accoun simply as he change in he sock of ne foeign asses, i.e.: NII NII 1 (2) CA = NFA NFA 1 =. We can fuhe undesand he souces of he sock of dak mae () by noicing ha: NII (3) = NFA NFA = NFA = ~ ( NFA + µ ) ~ ( ~ ) NFA = µ + NFA whee NFA sands fo he official measue of ne foeign asses. In his expession we allow fo asses o be mismeasued, wih µ indicaing ha eo in measuemen. In 3 This way of compuing ne foeign assess has been suggesed by Cline (2005) and peviously by Ulan and Dewald (1989). I was discussed by U.S. govenmen officials, bu he Bueau of Economic Analysis evenually discaded i due o he difficulies associaed wih choosing a discoun ae. (See Landefeld and Lawson, 1991) addiion, we assume asses yield a ae of eun ~ diffeen fom he consan ae used fo discouning. The wo ems in he las expession of equaion (3) allow visualizing ha dak mae may have wo oigins: he capialized eun o unaccouned asses and eun pivileges. These pivileges may aise because asses aboad ean a highe ae of eun o because liabiliies a home pay a lowe ae. We agee wih HKT ha i is ielevan fom he pespecive of ne income o say ha he asse base is lage (say, because, asses ae mismeasued) and eans a lowe yield, o ha he asse base is smalle and eans a highe eun. In ou oiginal wok we disinguish he wo channels because he undelying economics ae diffeen. Thus, alhough i is useful o useful idenify he wo channels sepaaely, o sudy he elevance of dak mae we can assume µ=0, associaing all he effec o he yield diffeenials. If so equaion (3) becomes: ( ~ ) (4) = NFA Equaion (4) gives he capialized value of he eun diffeenial elaive o a given benchmak ae. Once a discoun ae is chosen we can easily compue his meas- FIGURE 1 U.S. STOCK OF DARK MATTER (IN TRILLIONS OF U.S. DOLLARS, AND IN PERCENT OF GDP) 6 5 in illions of dollas (lef scale) in % of GDP (igh scale) 45% 40% 35% 4 30% 3 25% 20% 2 15% 1 10% 5% 0 0% 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Souce: Bueau of Economic Analysis The Valuaion of Hidden Asses in Foeign Tansacions: Why Dak Mae Maes Business Economics Januay 2007 31
ue fo any couny in he wold. Figue 1 shows he sock of dak mae asses fo he Unied Saes when a five pecen discoun ae is used. I has been gowing pey seadily, acceleaing owads he end of he peiod. On aveage he incease in dak mae asses has aveaged 2.4 pecen of GDP pe yea since 1982, bu close o five pecen of GDP in ecen yeas. Equaion (4) saes ha dak mae oiginaes in eun pivileges, ha is, highe elaive eun fo asses han fo liabiliies. We ague ha hee ae a leas hee main facos ha explain a pesisen eun diffeenial beween asses and liabiliies: a eun diffeenial fo FDI invesmens, he sale of insuance, and he povision of liquidiy sevices. The fis faco involves he noion ha FDI invesmens aboad ae a vehicle fo he disseminaion of inellecual popey such as ideas, bluepins, and knowledge, and ha hey ae he vehicle fo unaccouned expos of sevices poduced by headquaes and used by affiliaes aound he wold. FDI invesos puchase asses in ode o inves, bu also bing wih hem a bluepin, a poduc, and a business know-how ha is usually pooly accouned fo. Because of he difficulies in acking he ade of hese sevices acoss naional bodes, i is likely ha numbes undeesimae boh he ne woh of he companies o he expos of he souce counies ha ae hen used by foeign affiliaes o geneae income, all of which show up as a lage yield fo hese asses (See Higgins, Kligaad and Tille, 2005). The second faco may aise because he undelying sabiliy o insabiliy of a given economy may allow some economies o sell some of his sabiliy o he es of wold, and o do so hey will chage isk pemia and ean a seady flow fo hese insuance sevices. The hid faco is elaed o he povision of liquidiy sevices, eihe diecly hough he use he domesic cuency aboad, o because asses wih deep makes ean a liquidiy pemium. In Hausmann and Suzenegge (2006b), we use a coss secion of counies and discuss elaed lieaue poviding evidence fo all of hese facos. By capializing he income fom hese eun diffeenials we ae edefining he sock of asses in a way ha moe explicily shows he value of undelying capial and sevices, egadless of whehe hey ae included in official saisics. In ohe wods, capializing hese seady eun diffeenials allows including in he asse esimaes he advanage o disadvanage efleced in he diffeences in he aes of eun, hus poviding a moe coheen picue of acual global imbalances. As we will show below, assimilaing he eun diffeenial o an asse is anamoun o assuming ha his diffeenial will pesis in he fuue. Heein lies he main diffeence when looking a global imbalances fom he pespecive of dak mae. The ciique HKT cas doub on ou valuaion pocedue by using a simple agumen. Accep fo a momen ha i makes sense o discoun he invesmen income by, say, a consan discoun ae and specifically by he five pecen ae used in HS. If his is coec, one could apply he valuaion pocedue no only o he ne eun bu also sepaaely o he income obained fom asses and liabiliies. Howeve, when doing so wih U.S. daa, HKT find ha some asses and liabiliies ae undevalued elaive o official numbes. They ague his indicaes hee is dak animae ahe han dak mae in hese asses and liabiliies, and only because hee is lage dak animae in liabiliies does he Unied Saes appea as having dak mae asses. They claim his is evidence agains he exisence of dak mae in U.S. asses. We believe his misses he poin. The undelying famewok we use is simply o claim ha hee has o be a eason why asses delive ex pos diffeen ae of euns (we suggesed isk pemia, know-how, and liquidiy as poenial concepual explanaions). We define dak mae fo each asse as he capialized value of he deviaions in euns fom a benchmak ae, which measues he eun o hese hidden aibues (poduciviy, isk, o liquidiy). Fo example, he capialized value of he isk pemia of a sock eun will be is dak mae, and he liquidiy discoun on a financial insumen is anohe souce of dak mae. Thus valued, asses and liabiliies may conain posiive o negaive amouns of dak mae. Asses will conain posiive dak mae if hei eun is lage han a benchmak inees ae. Fo easonable aes hee is posiive dak mae in U.S. FDI asses, bu no necessaily so fo ohe asses. Liabiliies conain dak mae if hei eun is below he benchmak. In he case of he Unied Saes, his is paiculaly elevan fo deb, which pays low euns and has liquidiy advanages. In fac, finding ha hese asses ae woh less han make value is equivalen o confiming he exisence of dak mae in liabiliies, no he ohe way aound. HKT associae dak mae o know-how asses, and hus look fo i only wihin U.S. asses, bu his misses one of he key souces of dak mae fo he Unied Saes he fac ha insuance and liquidiy consideaions allow i o pay much lowe aes on is liabiliies. Gouinchas and Rey (2006) find i is in deb insumens whee he Unied Saes commands he highes goss eun diffeenial. This uns ou o be a ciical souce of dak mae fo he Unied Saes, hough no unique o i. Kugle and Wede (2004, 2005) show ha Swizeland benefis fom simila if no lage benefis. 32 Business Economics Januay 2007 The Valuaion of Hidden Asses in Foeign Tansacions: Why Dak Mae Maes
Of couse i may be asked whehe ou measue povides an alenaive o make values. We hink i does no. Bu ou poin emains ha i is useful o ake ino accoun sysemaic eun diffeenials when analyzing global imbalances because i is feasible ha hese asses will be deliveing seady eun diffeenials in ha fuue ha will be elevan o assess vulneabiliies. By focusing on make values o assess asse posiions his feaue is oally missed. 4 Why does i mae? Having claified ha finding a lowe value of U.S. liabiliies does no only no pose any poblem wih he pocedue used, bu on he conay is a he coe of he agumen, we ackle now he quesion of explaining why we believe ha dak mae povides an alenaive vanage poin fo undesanding cuen accoun imbalances. 4 The exac amoun of dak mae is abiay, because i depends on he abiaily chosen discoun ae. If he chosen ae inceases, we will find less dak mae in asses and moe of i in liabiliies. If he chosen ae deceases, we will find moe of i in asses and less in liabiliies. The ne effec, while less sensiive, will also depend on he discoun ae used (as an exeme example if we choose he evidenly uneasonable low ae of 0.1 pecen as a discoun faco, we would To illusae his, le us use a simple example ha oughly elaes o he cuen U.S. ne asse posiion. Imagine a couny wih foeign asses o GDP of 100 pecen and foeign liabiliies o GDP of 125 pecen. Jus fo he sake of he agumen, imagine ha asses yield a 6.25 pecen eun while liabiliies pay five pecen, hus yielding a balanced income flow in spie of measued ne foeign asses implying ne liabiliies of 25 pecen of GDP. Thee ae wo ways o inepe his esul. The convenional way is o say ha in his couny asses yield a elaively highe ae of eun ha compensaes fo he highe liabiliies. The dak mae inepeaion, on he ohe hand, would descibe his by sessing ha he inees diffeenial eflecs undelying fundamenals and heefoe should be inepeed as he eun o an unobsevable asse. In his example, asses yield 1.25 pecen descibe he Unied Saes as having ne liabiliies of seven illion U.S. dollas, way above is cuen value, esuling fom an asse valuaion of 585 illion and liabiliies of 592 illion i.e., measuing boh asses and liabiliies a close o 60 imes hei make value, hough all eaning he paly 0.1 pecen eun). While descibing he daa his way would delive viually he same message, we believe choosing a easonable ae of eun, close o isk-fee aes, povides a moe inuiive inepeaion. FIGURE 2 THE U.S. INTEREST SPREAD 1976-2005 10% 9% Implici Reun on Asses Implici Reen on Liabiliies Spead 8% 7% 6% 5% 4% 3% 2% 1% 0% 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Souce: Bueau of Economic Analysis The Valuaion of Hidden Asses in Foeign Tansacions: Why Dak Mae Maes Business Economics Januay 2007 33
moe han liabiliies, so hee ae implici asses equal o 25 pecen of GDP (1.25/.05 o sick o ou five pecen ae). Of couse he euns o he discoun faco can be changed, bu he undelying soy would be he same. Wha is he diffeence beween saing ha he Unied Saes eans a eun diffeenial on is asses and liabiliies, o ha i holds unmeasued asses in an amoun equivalen o he capialized value of he eun diffeenials elaive o a benchmak ae? We believe hee ae basically wo diffeences. The fis is ha by assigning asse-ype chaaceisics o hese eun diffeenials, we ae implicily assuming ha his eun diffeenial will pesis going fowad i.e., he null hypohesis becomes ha he undelying sock of asses and liabiliies will coninue o delive such eun diffeenials in he fuue. Noice ha in he adiional inepeaion, he null hypohesis is ha his will no happen, so ha an explanaion needs o be given evey peiod why he asses ae no inducing a commensuae ne income flow. 5 Bu wha evidence is hee ha hese euns ae elaively sable? In he case of he Unied Saes he evidence is faily compelling. Figue 2, shows he eun diffeenials compued fo asses and liabiliies fom official U.S. daa. I seems ha he null hypohesis ha hese euns will coninue o delive a eun diffeenial, a leas in he case of he Unied Saes, appeas pefecly easonable. The second eason fo he diffeence is ha he value of he unobsevable asses may be gowing ove ime. Tacking he sock of dak mae ove ime is a way of esimaing if his is indeed happening. Figue 1 depiced he evoluion of he value of hese asses fo he Unied Saes, in fac showing a gowing sock of dak mae. This inceasing sock of dak mae is pobably he combinaion of a faily consan yield diffeenial, shown in Figue 2, combined wih inceased leveage of his diffeenial as goss asse posiions inceased hand-in-hand wih an inceasingly inense pocess of financial globalizaion. So when pedicing he evoluion of U.S. cuen accoun imbalances in he fuue, should he analyses ignoe he fac ha he Unied Saes will coninue o mase a eun diffeenial and ha i may be able accumulae dak mae asses? If hee is evidence ove he las 5 HKT spend a gea deal of hei pape explaining measued valuaion effecs ha explain why he ne foeign asse posiion has no deeioaed hand-in-hand wih he cuen accoun. Noice, howeve, ha ou valuaion effecs ae lage, even han hose descibed by HKT, as in ou measuemen of ne foeign asses being significanly lage han BEA s measuemen. 25 yeas ha i has accumulaed dak mae asses a he une of 2.4 pecen pe yea (and close o five pecen ove he las five yeas), doesn i make sense o exapolae his end fowad? Analyss can have hei own view on whehe he U.S. economy will be able o coninue expoing dak mae (i.e. of susaining a eun pivilege on is ne foeign asses). Ou wok povides some hisoical facs ha may become guidance as o wha o expec fo he fuue. Sho as i is, he daa fo he las 25 yeas sugges ha i may make sense o assume ha hese inceases in he sock of dak mae will pesis ino he fuue. In fac, as he wold becomes inceasingly globalized, i is likely ha some of he souces of dak mae may be bound o incease. If so, he deeioaion of ne income will be significanly slowe han wha official numbes anicipae. In he end i boils down o he diffeence beween evaluaing he fuue pospec fo a couny wih ne foeign deb of abou 25 pecen of GDP and unning a cuen accoun defici of six pecen of GDP wih ha of a couny ha has no ne foeign liabiliies and uns a cuen accoun of beween one and 3.5 pecen of GDP 6. The diffeence beween hose wo scenaios povides he scope fo undesanding he elevance of he dak mae appoach o inepeing fuue U.S. and global imbalances. A a minimum, he dak mae hypohesis focuses he quesion of he U.S. cuen accoun imbalances on he undelying moives fo he eun diffeenials. Rahe han concenaing on savings behavio, he analysis suggess ha i is equally impoan o evaluae whehe he Unied Saes will coninue o be a safe haven fo invesmens, un a sable and aacive cuency, and have he mos innovaive univesiies and fims. These issues have been lagely absen fom he debae on U.S. cuen accoun imbalances, bu ou appoach focuses squaely on hem. REFERENCES Buie, Willem. 2006. Dak Mae o Cold Fusion? Goldman Sachs, Global Economics Pape No. 136. Caballeo, R., E. Fahi, and P. O. Gouinchas. 2005. An Equilibium Model of Global Imbalances and Low Inees Raes. Mimeo, MIT. Sepembe. Campbell, John, Kaine Sefay-de Medeios, and Luis Viceia. 2006. Global Cuency Hedging. Mimeo, Havad Univesiy. Chai, Anusha and Linda Tesa. 2006. The Make s Response o Developed-Make Acquisiions in Emeging Makes, Mimeo, Univesiy of Michigan. Cline, W. 2005. The Unied Saes as a Debo Naion. Insiue fo Inenaional Economics. Gouinchas P. O. and H. Rey. 2006. Fom Wold Banke o Wold 34 Business Economics Januay 2007 The Valuaion of Hidden Asses in Foeign Tansacions: Why Dak Mae Maes
Venue Capialis: U.S. Exenal Adjusmen and he Exobian Pivilege, in Claida, R.H. (Ed.). G-7 Cuen Accoun Imbalances: Susainabiliy and Adjusmen. Chicago IL: Univesiy of Chicago Pess. Hausmann, Ricado and Fedeico Suzenegge. 2005a. Dak Mae Makes he U.S. Defici Disappea. Financial Times. Decembe 8, p. A15. Hausmann, Ricado and Fedeico Suzenegge. 2006a. Why he U.S. Cuen Accoun is Susainable. Inenaional Finance. 9:2, pp. 223-240. Hausmann, Ricado and Fedeico Suzenegge. 2006b. U.S. and Global Imbalances: Can Dak Mae Peven a Big Bang? Woking pape, Kennedy School of Govenmen. Novembe. Higgins, Mahew, Thomas Kligaad, and Cedic Tille. 2005. The Income Implicaions of Rising U.S. Inenaional Liabiliies. Cuen Issues in Economics and Finance, Fedeal Reseve Bank of New Yok. 11:12. Higgins, Mahew, Thomas Kligaad and Cedic Tille. 2007. Boowing Wihou Deb? Undesanding he U.S. Inenaional Invesmen Posiion. Business Economics. Januay. Kugle, P. and B. Wede. 2004. Inenaional Pofolio Holdings and Swiss Fanc Reuns. Mime, Univesiy of Mainz. Kugle, P. and B. Wede. 2005. Why ae Reuns on Swiss Fanc Asses so Low? Rae Evens May Solve he Puzzle. Applied Economics Quaely. 51:3, pp. 231-246. Landefeld S. and A. Lawson. 1991. Valuaion of he U.S. Ne Inenaional Invesmen Posiion. Suvey of Cuen Business. May, pp. 40-49. Meissne, C. and A. Taylo. 2006. Losing ou Mables in he New Cenuy? The Gea Rebalancing in Hisoical Pespecive. Mimeo, London School of Economics. Mendoza, Enique, Vincenzo Quadini, and Jose Vico Rios-Rull. 2006. Financial Inegaion, Financial Deepness and Global Imbalances. Mimeo, Univesiy of Mayland. Ulan, M. and W. Dewald. 1989 The U.S. Ne Inenaional Invesmen Posiion: Missaed and Misundesood, in J. Don and W. Niskanen (Eds.) Dollas, Deficis and Tade. Kluwe Academic Publishes. The Valuaion of Hidden Asses in Foeign Tansacions: Why Dak Mae Maes Business Economics Januay 2007 35