1、American Economic Review 101 (February 2011): 196233http:/wwwnullaeawebnullornull/articnullenullnullphpnullnulloi=10null12nullnull/aernull101null1null196196Over the last 30 years, China has undergone a spectacular economic transforma-tion involving not only fast economic growth and sustained capital
2、 accumulation, but also major shifts in the sectoral composition of output, increased urbanization and a growing importance of markets and entrepreneurial skills. nullallocation of labor and capital across manufacturing nullms has been a key source of productivity growth. nulle rate of return on inv
3、estment has remained well above null percent, higher than in most industrialized and developing economies. null investment rates have been high, saving rates have been even highernullin the last null years, China has enullerienced a growing net foreign surplusnullits foreign reserves swelled from nu
4、ll billion nullnullin nullnull (nullpercent of its annual nullnull) to nullnull0 billion nullnullin nullne null0null (null percent of its nullnull)nullsee nullgure nullnulle combination of high growth and high return to capital, on the one hand, and a growing foreign surplus, on the other hand, is p
5、uzzling. nullclosed-economy Growing Like Chinanull nullnullnullnullnull, nullnullnull nullnullnullnullnullnull, nullnullnullnullnullnullnullnullnullnullnullnullnulle connulltruct a nullrowth monullenull connullinulltent with nullhinanullnull economic trannullition: hinullh output nullrowthnull nullu
6、nulltainenull returnnull on capitanullnull reanullnullonullcation within the manunullacturinnull nullectornull annull a nullarnulle tranulle nullurpnullunullnullEntrepreneurianull nullrmnull unulle more pronulluctive technonullonullienullnull but nullue to nullnancianull impernullectionnull they mun
7、ullt nullnance invenulltmentnull throunullh internanullnullavinnullnullnull nulltatenullownenull nullrmnull have nullow pronulluctivity but nullurvive becaunulle onull better accenullnull to crenullit marnulletnullnull nullinullhnullpronulluctivity nullrmnull outnullrow nullownullpronulluctivity nul
8、lrmnull inull entrepreneurnull have nullunullnullcientnully hinullh nullavinnullnullnullnullhe nullownnullinullinnull onull nullnancianullnully intenullratenull nullrmnull nullorcenull nullomenulltic nullavnullinnullnull to be invenulltenull abroanullnull nulleneratinnull a nulloreinulln nullurpnull
9、unullnull A canullibratenullvernullion onull the theory accountnull nulluantitativenully nullor nullhinanullnull economic trannullitionnull (nullEnull nullnullnull, nullnullnull, nullnull3, nullnull3, nullnull0, Onullnull, Onull3, nullnull3, nullnullnull, null3null) null nullngnull nullhool of nullo
10、nomics, nulldan nulliversity, nullnullnull0 nullonullan nullad, nullanghai null0null3, China and Chinese nulliversity of nullng nullong, nullnullnullther nulle nullilding, nullatin, nullng nullong (e-)nullnulloreslettennullnullsearch nullpartment, nullderal nullserve nullnk of nullnneapolis, null nu
11、llnnepin nullvenue, nullnneapolis, null nullnullnull (e-)nullnulllibottinullnullpartment of nullonomics, nulliversity of nullrich, nullhlebachstrasse null, Cnullnull0nullnullrich (e-mailnullfabrizio.zilibottinullcon.uzh.ch). nulle thank the two ref-erees, Chong-null nulli, nullancesco Caselli, Chang
12、-nullai nullieh, nullikai nullang, and nullaodong nullu for comments. nulle also thank participants at the nullnullconference on nullhe nullergence of China nullnulldia in the nullobal nullonomynull(nullly 3null, null0null), the China nullonomics nullmmer nullstitute null0null the null0nullnullnneso
13、ta nullorkshop in nullcroeconomic nulleory, and the null0nullnullinghua nullorkshop in nullcroeconomics, and seminar participants at various departmental seminars. nullnancial support from the nullropean nullsearch Council (nullC nullvanced nullant nullCnull-nullnullnull), from the nulliss nulltiona
14、l nullience nulloundation (null00null-nullnull3null), from the nullsearch Council of nullrway (grants nullnullnull and null3nullnull), from the nullnullplatform at nulldan, the nullcial nullience nulloundation of China (0nullnull00null), and the nulltural nullience nulloundation of China (nullnull30
15、0null) are gratefully acknowledged. nullis paper is part of the research activities at nullOnullsupported by the nullsearch Council of nullrway (nullnullnull). nully views enullressed here are those of the authors and not those of the nullderal nullserve nullnk of nullnneapolis or the nullderal null
16、serve nullstem.19nullnullnullnullnull Enull Anullnull: nullRnullnullnullnullnull nullnullnullE nullnullnullnullAnullnullnullnull 101 nullnullnull 1neoclassical growth model predicts that the high investment rate would lead to a fall in the return to capital. null open-economy model predicts a large
17、net capital innullw rather than an outnullw, owing to the high domestic return to capital. null this paper, we propose a theory of economic transition that solves this puzzle while being con-sistent with salient nullalitative and nullantitative features of the Chinese enullerience. nulle focal point
18、s of the theory are nullancial frictions and reallocation of resources across nullms. null our theory, both the sustained return to capital and the foreign sur -plus arise from the reallocation of capital and labor from less productive enullernally nullanced nullms to entrepreneurial nullms that are
19、 more productive but have less access to enullernal nullancing. null nullancially integrated nullms shrink, a larger proportion of the domestic savings is invested in foreign assets. nullus, the combination of high growth and high investment is consistent with the accumulation of a foreign surplus.O
20、ur paper is part of a recent literature arguing that low aggregate total factor pro-ductivity (nullnull)nullspecially in developing countriesnulls the result of micro-level resource misallocation (see nullephen null nullarente, nullchard nullgerson, and nullndall nullight null00nullnullancesco Casel
21、li and nullilbur null Coleman null null0null nullhijit nullnerjee and nullther nullnull null0null nullego nullstuccia and nullgerson null0null nullno nullncia and nullabrizio nulllibotti null0null and Chang-nullai nullieh and nullter null nullenow null0null). nullile pockets of efnullient nullms usi
22、ng state-of-the-art technologies may enullst, these nullms fail to attract the large share of productive resources that efnulliency would dictate, due to nullancial frictions and other imperfections. nullst of the enullsting literature 1992 1994 1996 1998 2000 2002 2004 20060102030405060YearPercenta
23、ge of GDPForeign reservesDifference between deposits and loansnullnullnullnullnullnull null. nullnullnullnullnullnullnull nullnullnullnullnullnullnullnull nullnullnull nullnullnull nullnullnullnullnullnullnullnullnullnull nullnullnullnullnullnullnull nullnullnullnullnullnullnullnull nullnullnull nul
24、lnullnullnullnullnullote: nulle nullure plots Chinanulls foreign reserves (solid line) and the domestic bank deposits minus domestic loans (dotted line), both enullressed as a percentage of nullnull.nullurce: Cnull, various issues.nullnullE AnullERnullnullAnull Enullnullnullnullnullnullnull REnullnu
25、llEnull19null FEnullRnullARnull 2011emphasizes the effects of resource misallocation on average productivity. null con-trast, our paper argues that when a country starts from a situation of severe inefnull-ciency but manages to ignite the engine of reallocation, it has the potential to grow fast ove
26、r a prolonged transition, since efnullient nullms can count on a highly elastic supply of factors attracted from the less productive nullms.nullo analyze such a transition, we construct a model in which nullrms are heteroge-neous in productivity and access to nullnancial markets. nulligh-productivit
27、y nullrms are operated by agents with entrepreneurial skills who are nullnancially constrained and who must rely on retained earnings to nullnance their investments. nullow-productivity nullrms can survive due to their better access to credit markets, since the growth poten-tial of high-productivity
28、 nullrms is limited by the enulltent of entrepreneurial savings. nullf the saving nullow is sufnullciently large, high-productivity nullrms outgrow low-produc-tivity ones, progressively driving them out of the market. nulluring the transition, the dynamic enulluilibrium has nullnull featuresnull wit
29、hin each type of nullrm, the rate of return to capital is constant due to labor mobility and to the nullnancial integration of the low-productivity nullrms. nullue to a composition effect, the aggregate rate of return to capital actually increases. nulloreover, the economy accumulates a foreign surp
30、lus. nullhile investments in the enullpanding nullrms are nullnanced by the retained earnings of entrepreneurs, wage earners deposit their savings with intermediaries who can invest them in loans to domestic nullrms and in foreign bonds. nulls the demand for funds from nullnancially integrated domes
31、tic nullrms declines, a growing share of the intermediated funds must be invested abroad, building a growing foreign surplus. nullhis prediction is consistent with the observation that the difference between deposits and domestic bank loans has been growing substantially, tracking Chinanulls accumul
32、ation of foreign reserves (see again nulligure null). nullfter the transition, the economy behaves as in a stan-dard neoclassical model, where capital accumulation is subject to decreasing returns.nullallocation within the manufacturing sectornullhe driving force in our modelnullhas been shown to be
33、 an important source of productivity growth in China. null an innullential paper, nullieh and nullenow (null0null) estimate that reallocation across manu-facturing nullms with different productivity accounted for an annual two percentage point increase in aggregate nullnullduring nullnullnull00null
34、nullren nullandt, nullhannes nullan nullesebroeck, and nullifan nullang (null0null) estimate that up to two-thirds of the aggregate nullnullgrowth in Chinese manufacturing was due to productivity differences between entering and enullting nullms during nullnullnull00nullOur theory yields several add
35、itional predictions consistent with the evidence of Chinanulls transitionnull(i) nulle theory predicts that the surplusnullavings minus investmentnullhould increase with the share of entrepreneurial nullms. Consistent with this predic-tion, we nulld that the net surplus is signinullantly higher in C
36、hinese provinces in which the employment share of domestic private nullms has increased faster.(ii) null our benchmark model, all nullms produce the same good and differ only in nullnull. nulle enullend the theory to a two-sector model in which nullms can special-ize in the production of more or les
37、s capital-intensive goods. nullis enullended model predicts that nullancially constrained nullms with high nullnullwill spe-cialize in labor-intensive activities (even though they have no technological 199nullnullnullnull Enull Anullnull: nullRnullnullnullnullnull nullnullnullE nullnullnullnullAnull
38、nullnullnull 101 nullnullnull 1comparative advantage). nullus, the transition proceeds in stagesnullnullst low-productivity nullms retreat into capital-intensive industries, and then they gradually vanish. nullis is consistent with the observed dynamics of sectoral reallocation in China, where young
39、 high-productivity private nullms have entered enullensively in labor-intensive sectors, while old state-owned nullms continue to dominate capital-intensive industries.nulle theory is related to the seminal contribution of nullthur null. nullwis (nullnull), who constructs a model of reallocation fro
40、m agriculture to industry where the supply of labor in manufacturing is unlimited due to structural overemployment in agriculture. nullile his mechanism is similar in some respects to ours, productivity increases in his model rely on some form of hidden unemployment in the traditional sector. nullwi
41、snulltheory captures aspects of the reallocation between rural and urban areas in China, while our focus is on the reallocation within the industrial sector. Our paper is also related to nullume nullentura (nullnull), who shows that in economies engaging in enullernal trade, capital accumulation is
42、not subject to diminishing returns because resources are moved from labor-intensive to capital-intensive sectors. nullenturanulls model does not assume any initial inefnulliency, nor does it imply that nullnullshould grow within each industrynull key implication of our theory.nullnullither nullwisnu
43、llnor nullenturanulls theory has any implication regarding trade imbal-ances. nullminori nulltsuyama (null0null null0null) shows that nullancial frictions may induce trading economies to specialize in industries in which they do not have a technologi-cal comparative advantage. nulle also the work of
44、 nulll nulltrnull and nullcardo null Caballero (null0null). null our model, by a similar mechanism, less efnullient nullms can survive and even outgrow more productive ones. Our two-sector enullension also predicts that nullancial constraints generate specialization in spite of the lack of any techn
45、ological comparative advantage, though the mechanism is different.nullerre-Olivier nullurinchas and Olivier nullanne (null0null) document that it is common to observe capital outnullw from fast-growing developing economies with high mar -ginal product of capital. null in the case of China, countries
46、 with fast nullnullgrowth tend to have both large capital outnullws and large investment rates, while the opposite is true for slow-growing countries. nulley label this nullding the nullllocation puzzle.nullOur theory can provide a rationale to this observation. null a related paper, nullancisco nul
47、l nullera and nullongseok nullin (nullnull) focus on the current account surpluses enulleri-enced by a number of nullian economies in the nullnulls (with the notable enulleption of China, which enullerienced current account denullits during the nullnulls). nullera and nullin arguenulls we donullhat
48、nullancial frictions can contribute to the enulllanation of this puzzle. nullile in our paper the foreign surplus is driven by the dwindling demand for domestic borrowing, due to the decline of nullancially integrated nullms, they emphasize increased domestic savings by agents who are planning to be
49、come entrepreneurs but need to save to nullance start-up costs.nullfew recent papers address the more specinull nullestion of why China is accumu-lating a large foreign surplus. nullst papers emphasize the countrynulls high saving rate. nulluis nulluijs (null0null) shows that household and enterprise saving rates in China are, null null this respect, our work is related to the seminal papers of nullmon nulluznets (nullnull) and nullllis Chenery and nullises nullrnullin (nullnull), who study sources of productivity