Dis­tor­tions, Growth Catch-up, and Sus­tain­able Growth

China Economist - - Articles - ZhangXiao­jing(张晓晶),LiCheng(李成)andLiYu(李育)

Ab­stract: A dis­tor­tion is a de­par­ture from the al­lo­ca­tion of eco­nomic re­sources from the state in which each agent max­i­mizes his/her own wel­fare. Dis­tor­tions can be di­vided into en­doge­nous dis­tor­tion (i.e. mar­ket im­per­fec­tions) and pol­icy-im­posed dis­tor­tion. The re­la­tion­ship be­tween dis­tor­tion and de­vel­op­ment is com­plex, thus fa­vor­able dis­tor­tion would only be pos­si­ble un­der cer­tain con­di­tions, where, as ar­gued in this pa­per, four cru­cial mech­a­nisms may play roles — ad­van­tage of back­ward­ness, sec­ond-best prin­ci­ple, co­or­di­na­tion fail­ure and po­lit­i­cal econ­omy per­spec­tive. Em­pir­i­cally, both in­ter­na­tional ex­pe­ri­ence and ev­i­dence from China sug­gest that dis­tor­tions have a pos­i­tive ef­fect on to­tal fac­tor pro­duc­tiv­ity (TFP) in the early stages of de­vel­op­ment, but with in­creas­ing in­come lev­els this role grad­u­ally di­min­ishes. Es­pe­cially in the phases of mid­dle and high in­come, the neg­a­tive ef­fects of dis­tor­tions are sig­nif­i­cant and be­come an im­por­tant fac­tor lead­ing to the mid­dle-in­come trap. There­fore, re­duc­ing and cor­rect­ing dis­tor­tions is the key to achiev­ing sus­tain­able growth. Re­gard­ing China, it is nec­es­sary to elim­i­nate the dis­tor­tions in a clear way and let the mar­ket play the de­ci­sive role in re­source al­lo­ca­tion. Oth­er­wise, in the name of “growth catch-up,” the pol­icy-im­posed dis­tor­tion will oc­cur fre­quently, and the di­rec­tion of mar­ket-ori­ented re­form will be­come blurred and swing. Mit­i­gat­ing un­fa­vor­able dis­tor­tions is largely a process of ex­plor­ing the fa­vor­able bor­der­line of gov­ern­ment and mar­ket, which con­sti­tutes a ma­jor chal­lenge for all economies.

Key­words: dis­tor­tions, growth catch-up, gov­ern­ment–mar­ket re­la­tion­ship, sus­tain­able

growth

JEL Clas­si­fi­ca­tion Codes: D23; H1; O1; O2 DOI: 1 0.19602/j .chi­nae­conomist.2018.09.02

1. In­tro­duc­tion

The ques­tion of whether rapid eco­nomic growth is nec­es­sar­ily fol­lowed by slow­down and stag­na­tion has been at the cen­ter of re­cent aca­demic de­bate (Eichen­green et al. 2011; Pritch­ett and Sum­mers 2014). It can be fur­ther di­vided into some sub-ques­tions, in­clud­ing what are the driv­ing forces of rapid growth, what are the chang­ing roles played by mar­ket re­forms, fac­tor ac­cu­mu­la­tions (es­pe­cially in­vest­ment and la­bor force) and gov­ern­ment be­hav­ior in the course of de­vel­op­ment, and what are the key fac­tors to achieve sus­tain­able de­vel­op­ment. Un­doubt­edly, these ques­tions sorely need an­swer­ing in the case

of present- day China, which is chal­lenged by eco­nomic slow­down and so­cioe­co­nomic struc­tural trans­for­ma­tions. This pa­per at­tempts to shed some light on this is­sue by fo­cus­ing on “dis­tor­tion,” a fac­tor of both the­o­ret­i­cal and prac­ti­cal im­por­tance.

From a the­o­ret­i­cal per­spec­tive, a “dis­tor­tion” refers to a de­par­ture from the per­fect com­pet­i­tive equi­lib­rium with no ex­ter­nal­i­ties in which re­sources have been op­ti­mally al­lo­cated so that each eco­nomic agent max­i­mizes his/her own wel­fare. It is usu­ally caused by mar­ket im­per­fec­tion. In the above sense, dis­tor­tion is per­va­sive in economies of all in­come lev­els, but their char­ac­ter­is­tics, ex­tents, types, forms and im­pacts on so­cioe­co­nomic de­vel­op­ment dif­fer sub­stan­tially across coun­tries.

In par­tic­u­lar, judged from their causes, dis­tor­tions can be di­vided into two cat­e­gories. The first is “en­doge­nous dis­tor­tion,” caused by mar­ket im­per­fec­tion and un­der­de­vel­op­ment. The sec­ond is “pol­i­cy­im­posed dis­tor­tion,” caused by gov­ern­ment poli­cies and in­ter­ven­tions. The lat­ter is of­ten founded upon or even on the pre­text of the for­mer (Bhag­wati 1969).

There are, in the­ory, two ex­treme types of dis­tor­tions. The “night-watch­man state” where econ­omy func­tions un­der the lais­sez-faire prin­ci­ple with min­i­mal dis­tor­tions, and “pa­ter­nal­ist state” whose sys­tem of wel­fare cov­ers from “cra­dle to grave,” which are com­monly con­sid­ered as two ex­treme types. Most coun­tries, if not all, lie in be­tween them.

Judged from the forms, there are more in­di­rect and highly in­sti­tu­tion­al­ized dis­tor­tions, with the aim of ad­just­ing mar­ket fail­ure and of pro­vid­ing pub­lic ser­vices of pos­i­tive ex­ter­nal­ity, such as ele­men­tary ed­u­ca­tion, en­vi­ron­men­tal pro­tec­tion and ba­sic re­search. Some are more di­rect and of ad­min­is­tra­tive color, with gov­ern­ments di­rectly in­volved in eco­nomic con­struc­tions and in­dus­trial de­vel­op­ment, usu­ally through state-owned en­ter­prises and se­lec­tive in­dus­trial poli­cies. It should be stressed that di­rect dis­tor­tions are of­ten associated with strate­gies of catch­ing-up. The lat­ter could, how­ever, back­fire in prac­tice, es­pe­cially when na­tional con­di­tions and fac­tors re­lated to the de­vel­op­ment phase are dis­re­garded.

Judged by im­pact on eco­nomic per­for­mance, dis­tor­tions dif­fer in many as­pects across dif­fer­ent coun­tries. Gen­er­ally speak­ing, at the early stage of de­vel­op­ment, fac­ing do­mes­tic in­sti­tu­tional im­per­fec­tion and heated com­pe­ti­tion from abroad, in­ten­sive dis­tor­tions, es­pe­cially se­lec­tive in­dus­trial poli­cies and de­vel­op­ment strate­gies, can be used for mo­bi­liz­ing re­sources, ac­cu­mu­lat­ing cap­i­tal and pro­tect­ing “in­fant in­dus­tries.” Nev­er­the­less, as an econ­omy be­comes more ma­ture in terms of both in­dus­trial struc­ture and mar­ket in­sti­tu­tions, dis­tor­tions’ ad­verse ef­fects on the econ­omy be­gin to man­i­fest them­selves. In par­tic­u­lar, through rent seek­ing by the gov­ern­ment sec­tor, dis­tor­tions may change the in­cen­tives and be­hav­ior of pri­vate eco­nomic agents, thereby hin­der­ing in­no­va­tion and long-term growth. In sum­mary, the ques­tion of what the “nec­es­sary” dis­tor­tions are in the con­text of growth catch-up

1 re­mains up for dis­cus­sion. To a large ex­tent, it per­tains to the choice of China’s de­vel­op­ment path in the fu­ture. With­out a solid the­o­ret­i­cal un­der­stand­ing of dis­tor­tion, there is a risk that re­forms will fal­ter or be blindly pur­sued, thus trap­ping the coun­try at the mid­dle-in­come level.

This pa­per at­tempts to pro­vide a com­pre­hen­sive per­spec­tive on the re­la­tion­ship be­tween dis­tor­tions and eco­nomic per­for­mance, and to dis­cuss the pol­icy im­pli­ca­tions of dis­tor­tions for China’s re­forms and sus­tain­able de­vel­op­ment. The rest of the pa­per is or­ga­nized as fol­lows. Sec­tion 2 re­views the re­cent lit­er­a­ture on dis­tor­tions. Sec­tion 3 dis­cusses the fa­vor­able dis­tor­tions and their ap­pli­ca­ble con­di­tions. Sec­tion 4 pro­vides ev­i­dence from in­ter­na­tional ex­pe­ri­ences for the im­pact of dis­tor­tions on eco­nomic per­for­mance. Sec­tion 5 turns to China’s case, with the em­pha­sis on the mar­ke­ti­za­tion process of Chi­nese

re­gions. The last sec­tion con­cludes the pa­per and pro­vides pol­icy dis­cus­sions.

2. Lit­er­a­ture Re­view

Re­gard­ing aca­demic lit­er­a­ture, dis­tor­tion as a re­search topic was firstly pro­posed by Bhag­wati (1969),in which he gen­er­al­ized the the­ory of dis­tor­tions (mostly in trade) and wel­fare and pro­posed a con­cep­tual dis­tinc­tion be­tween “en­doge­nous” dis­tor­tions and “pol­icy-im­posed” dis­tor­tions.

Since then, dis­tor­tions have re­ceived at­ten­tion, but mainly on the ex­pe­ri­ences of de­vel­op­ing coun­tries, in­clud­ing tax­a­tion, state-owned en­ter­prises, ad­min­is­tra­tive monopoly, trade and in­dus­trial poli­cies, fi­nan­cial re­pres­sion, ex­change rate man­age­ment and gov­ern­ment reg­u­la­tions. In par­tic­u­lar, eco­nom­ics of de­vel­op­ment and in­sti­tu­tional eco­nom­ics have pro­vided in­ten­sive the­o­ret­i­cal anal­y­sis and pol­icy dis­cus­sions on eco­nomic take-off, and de­vel­op­ment strate­gies in de­vel­op­ing coun­tries, such as McKin­non (1973),Sah and Stiglitz (1987), Gross­man and Help­man (1994), Parker (1995), Sh­leifer and Vishny (1998), Hall and Jones (1999), Qian (2000), Gor­don and Li (2009) and Ace­moglu and Robin­son (2012). Rel­a­tively speak­ing, thanks to ma­tured mar­ket sys­tem, en­doge­nous dis­tor­tions are less per­va­sive in ad­vanced economies, where dis­tor­tions are mainly to do with tax­a­tion and gov­ern­ment dis­cre­tionary sta­bi­liz­ing poli­cies as doc­u­mented in Judd(1985),Cham­ley(1986),Lu­cas(1990)and Easterly(1993).

More in­ter­est­ingly, as China reaches the mid­dle-in­come level, its eco­nomic struc­tures, com­par­a­tive ad­van­tages, sources and path of long-term growth have all been sub­ject to sub­stan­tial changes. Thus, the re­la­tion­ship be­tween gov­ern­ment and mar­ket in gen­eral, and the mixed im­pact of dis­tor­tions on the econ­omy in par­tic­u­lar, have been at the cen­ter of re­cent aca­demic dis­cus­sions, in­clud­ing Parker (1995), Young (2000), Lin (2003), Zhu et al. (2005), Zhang (2008), Zhang (2005), Hsieh & Klenow (2009), Zhang and Cheng (2010), Li and Lin (2011), Song et al. (2011), Yan and Hu (2013) and Qian (2016). The top­ics of in­ter­est in­clude fac­tor price dis­tor­tions, tax­a­tion dis­tor­tions, fi­nan­cial re­pres­sion, frag­men­ta­tion of do­mes­tic mar­kets and dis­tor­tions re­gard­ing open­ness. Re­cently, Yang (2016) fur­ther stressed that mit­i­gat­ing the dis­tor­tions on fac­tor al­lo­ca­tion is the key to the struc­tural re­forms in to­day’s China.

It should be em­pha­sized that although some stud­ies do not di­rectly ad­dress the “dis­tor­tions,” they fo­cus on some ques­tions of high rel­e­vance, es­pe­cially the afore­men­tioned “pol­icy-im­posed” dis­tor­tions. In­deed, the very na­ture of the dis­tor­tion should be un­der­stood in the con­text of the gov­ern­ment–mar­ket re­la­tion­ship and the role of gov­ern­ment in de­vel­op­ment.

From this an­gle, early lit­er­a­ture on struc­tural­ist eco­nom­ics of de­vel­op­ment fo­cuses on mar­ket im­per­fec­tion in de­vel­op­ing coun­tries, and claims that the hand of gov­ern­ment can help re­pair the de­fi­cien­cies of the mar­ket. In other words, be­cause of the pres­ence of “en­doge­nous dis­tor­tions” (in­clud­ing mar­ket rigid­ity, hys­tere­sis, short­age over-sup­ply, in­elas­tic­ity of sup­ply and de­mand, and un­der­shoot­ing), gov­ern­ment needs to proac­tively in­ter­vene in the func­tion­ing of mar­kets through dis­tort­ing poli­cies and mea­sures, in­clud­ing forced sav­ings, “big-push,” in­dus­try se­lec­tion and im­port sub­sti­tu­tion. De­spite their pos­i­tive ef­fects in some ar­eas, the fo­cus on the gov­ern­ment in­ter­ven­tions brings about var­i­ous prob­lems, and thus it is re­placed by Neo­clas­si­cal eco­nom­ics of de­vel­op­ment, which em­pha­sizes the forces of mar­kets and the im­por­tance of price mech­a­nism. In fact, the story is more com­plex in prac­tice than in the­ory. Although some East Asian economies (such as Ja­pan and South Korea) avoided the “mid­dle in­come trap” thanks to, to a large de­gree, the ef­forts of gov­ern­ment, some fol­low­ers, such as Malaysia and In­done­sia, seem to have be­come trapped at the mid­dle-in­come level in the af­ter­math of the Asian Fi­nan­cial Cri­sis in the late 1990s. Later on, in­ten­sive dis­cus­sions and hot de­bates emerged around the so-called “Washington Con­sen­sus” ver­sus the “Bei­jing Con­sen­sus.” The un­der­ly­ing ques­tion of those con­cerns is once again that of the re­la­tion­ship be­tween gov­ern­ment and mar­ket. Of course, the topic be­came even more rel­e­vant in the wake of the 2008 cri­sis and a grow­ing body of lit­er­a­ture has emerged since then.

In his the­o­ret­i­cal frame­work la­beled New Struc­tural Eco­nom­ics, Lin(2014)stressed that an econ­omy should spe­cial­ize ac­cord­ing to its com­par­a­tive ad­van­tages and this path of de­vel­op­ment can­not be reached through a lais­sez-faire mar­ket mech­a­nism and thus, the ef­forts of gov­ern­ment are needed. Since com­par­a­tive ad­van­tages are mainly based on fac­tor en­dow­ment, in­clud­ing cap­i­tal, la­bor force and nat­u­ral re­sources, the changes in the lat­ter can also cause the changes in com­par­a­tive ad­van­tages. In­deed, to ex­ploit the com­par­a­tive ad­van­tages, both mar­kets and gov­ern­ment are indispensable. In this re­gard, two cases can be con­sid­ered: (1) If the gov­ern­ment gives pri­or­ity to the de­vel­op­ment of the in­dus­try of no com­par­a­tive and com­pet­i­tive ad­van­tages, the en­ter­prises in this in­dus­try will need pro­tec­tion ( i. e. a dis­tor­tion) be­cause of their lack of vi­a­bil­ity; ( 2) If the in­dus­try has com­par­a­tive ad­van­tages, the en­ter­prises in­volved will have high in­ter­na­tional com­pet­i­tive abil­ity thanks to their vi­a­bil­ity, thereby im­prov­ing their cap­i­tal ac­cu­mu­la­tions. As Lin ar­gued, gov­ern­ment should proac­tively im­prove the in­dus­trial up­grad­ing and tech­ni­cal in­no­va­tions through “growth iden­ti­fi­ca­tion” and “fa­cil­i­ta­tion.” In essence, the New Struc­tural Eco­nom­ics de­vel­oped by Lin at­tempts to strike a bal­ance be­tween gov­ern­ment and mar­ket. It seems that the for­mer has been overem­pha­sized by early struc­tural de­vel­op­ment eco­nom­ics and the lat­ter by the Washington Con­sen­sus.

In crit­i­ciz­ing the tra­di­tional the­ory of com­par­a­tive ad­van­tages, Stiglitz (2017) em­pha­sized the im­por­tance of learn­ing. He ar­gued that in the tra­di­tional the­o­ret­i­cal frame­work, com­par­a­tive ad­van­tage re­lies on the as­sump­tion that knowl­edge is pub­licly avail­able and its fo­cus is on fac­tor en­dow­ment (such as the cap­i­tal–la­bor ra­tio). Nev­er­the­less, since cap­i­tal is mo­bile across coun­tries, cap­i­tal en­dow­ment seems to be un­help­ful for un­der­stand­ing static com­par­a­tive ad­van­tage. In­stead, com­par­a­tive ad­van­tage is mainly de­ter­mined by im­mo­bile fac­tors, such as knowl­edge, la­bor and in­sti­tu­tions. The most im­por­tant of these is learn­ing abil­ity at the level of so­ci­ety. Given the fact that the mar­ket seems to be in­ef­fi­cient in pro­duc­ing and spread­ing knowl­edge, gov­ern­ments, es­pe­cially those in de­vel­op­ing coun­tries, need to play a proac­tive role in im­prov­ing learn­ing abil­ity.

From a broader per­spec­tive, Bard­han ( 2016) ad­dressed the roles of gov­ern­ment and their com­plex­ity. He ar­gues that be­cause of the com­pre­hen­sive­ness of the goals of de­vel­op­ment and the mul­ti­di­men­sion­al­ity of gov­ern­ment func­tions, along with co­or­di­na­tion fail­ure and the dif­fi­culty in col­lec­tive ac­tion, the role of gov­ern­ment needs redefining. From both his­tor­i­cal and log­i­cal per­spec­tives, Bard­han re­lated stages of de­vel­op­ment and gov­ern­ment’s roles. Jia (2011) and Wen (2016) also stressed the fol­low­ing fact: Gov­ern­ment in­ter­ven­tions played an im­por­tant and even indispensable role in the suc­cess of the early in­dus­tri­al­ized coun­tries.

Fukuyama(2012,2015)stressed the im­por­tance of “state ca­pac­ity,” and thus pro­vided an ar­gu­ment in sup­port of strong gov­ern­ment. He claimed that the pros­per­ity of a coun­try needs state ca­pac­ity, ac­count­able gov­ern­ment (democ­racy) and rule of law. Ac­cord­ing to him, the United States 1956) im­plies that if it is in, is suf­fer­ing from weak state ca­pac­ity, but en­joys ac­count­able gov­ern­ment and rule of law; in con­trast, China has strong state ca­pac­ity, but with weak or im­ma­ture ac­count­able gov­ern­ment and rule of law. Bes­ley and Pers­son(2009)also doc­u­mented that ex­pe­ri­ences in ad­vanced economies show that tax­a­tion and con­tract en­force­ment, as two im­por­tant el­e­ments of state ca­pac­ity, con­sti­tute nec­es­sary pre­con­di­tions for pros­per­ity. Nev­er­the­less, it should be noted that there is no stan­dard def­i­ni­tion for the term “state ca­pac­ity.” Loosely speak­ing, in our view, it in­cludes mar­ket sup­port (prop­erty rights de­lim­i­ta­tion and pro­tec­tion, con­tract en­force­ment, mar­ket rules), re­sources al­lo­ca­tion (such as tax­a­tion) and ad­just­ing so­cial re­la­tion­ships.

Ace­moglu et al. (2015) also ar­gued that the state ca­pac­ity is a cru­cial ex­plana­tory fac­tor for both the Asian Mir­a­cle and the medi­ocre growth per­for­mances of many African and South Amer­i­can coun­tries. This claim is fur­ther sup­ported by cross-coun­try data (Gen­naioli and Rainer 2007) and sub­na­tional data

Michalopou­los and Pa­paioan­nou, 2013;Bandy­opad­hyay and Green, 2012). It is note­wor­thy that the im­por­tance of state ca­pac­ity is by no means un­con­di­tional. It should be con­strained by ac­count­able gov­ern­ment and rule of law. With­out those two con­straints, a strong gov­ern­ment might turn into its

op­po­site, the Le­viathan, or grab­bing-hand gov­ern­ment. Thus, a strong gov­ern­ment should also be a lim­ited gov­ern­ment.

His­tor­i­cally, there has been a rise and fall in the gov­ern­ment’s role in eco­nomic de­vel­op­ment. In par­tic­u­lar, dur­ing eco­nomic booms, it is not un­com­mon that their role is ei­ther ig­nored or con­sid­ered as an im­ped­i­ment to pros­per­ity. In con­trast, dur­ing cri­sis their role is of­ten overem­pha­sized. The 2008 cri­sis is not an ex­cep­tion. In­deed, in the wake of this event, there was a grow­ing body of lit­er­a­ture fo­cus­ing on the (mainly pos­i­tive) role of gov­ern­ment. In this re­gard, four points should be made: (1) Gov­ern­ment, as a last re­sort, is able to tackle a cri­sis and mit­i­gate var­i­ous shocks, through sta­bi­liz­ing poli­cies such as debt-trans­mis­sion from the pri­vate sec­tor to the pub­lic sec­tor, es­pe­cially in ad­vanced economies; ( 2) Gov­ern­ment helps with in­no­va­tion. For ex­am­ple, Maz­zu­cato ( 2013) showed that gov­ern­ment in ad­vanced coun­tries ac­tu­ally plays a cru­cial role in in­no­va­tion through in­dus­trial poli­cies; (3) Gov­ern­ment, as a ven­ture cap­i­tal­ist, can also fill the gap be­tween pub­lic and pri­vate in­vest­ment; (4) Gov­ern­ment can not only make up for mar­ket de­fi­ciency, but proac­tively cre­ate the mar­ket as well. This

2 ar­gu­ment is in sharp con­trast with some in Neo­clas­si­cal eco­nom­ics, and thus has led to some crit­i­cism Min­gardi 2015).

It should be stressed that gov­ern­ment in­ter­ven­tions and dis­tor­tions are not in­ter­change­able. In other words, not all gov­ern­ment in­ter­ven­tion is seen as dis­tor­tion. More rig­or­ously, dis­tor­tions can be de­fined as gov­ern­ment in­ter­ven­tion that leads to the de­vi­a­tion from op­ti­mal al­lo­ca­tion of re­sources. In this spirit, the “mar­ket-aug­ment­ing gov­ern­ment” of Ol­son (2000) and the “mar­ket-en­hanc­ing gov­ern­ment” of Aoki (1997) should not be con­sid­ered as dis­tor­tions. Ac­cord­ing to Ol­son, the suc­cess of eco­nomic de­vel­op­ment should be based on two con­di­tions: One is the clear de­lim­i­ta­tion of rights, an­other is the pro­tec­tion of the econ­omy from the ex­tor­tion of pub­lic power. The “mar­ket-aug­ment­ing” gov­ern­ment is in­deed de­fined by these con­di­tions, which, in much the same spirit of “state ca­pac­ity,” im­ply that the gov­ern­ment should play a role in mar­ket cre­ation and func­tion. When dis­cussing gov­ern­ment in Eastern Asian economies, Aoki and oth­ers pro­posed the con­cept of “mar­ket-en­hanc­ing.” He ar­gued that be­cause gov­ern­ment fail­ure is not less per­va­sive than mar­ket fail­ure in eco­nomic ac­tiv­i­ties, var­i­ous in­dus­trial and so­cial or­ga­ni­za­tions, such as as­so­ci­a­tions of en­ter­prises, trade unions, fi­nan­cial in­ter­me­di­aries, and Cham­bers of Com­merce, can con­trib­ute to ad­dress­ing these two kinds of fail­ure. Thus, gov­ern­ment should also foster the de­vel­op­ment of these or­ga­ni­za­tions and build in­sti­tu­tional frame­works to co­or­di­nate with them. To judge whether a gov­ern­ment in­ter­ven­tion is a kind of dis­tor­tion, there­fore, the key cri­te­rion is whether this in­ter­ven­tion can re­in­force the mar­ket mech­a­nism.

In sum, like gov­ern­ment, dis­tor­tions play mixed roles in de­vel­op­ment, which de­pend on a num­ber of so­cioe­co­nomic char­ac­ter­is­tics and in­sti­tu­tional fac­tors. Given such com­plex­ity, it is very im­por­tant to in­ves­ti­gate, from both the­o­ret­i­cal and em­pir­i­cal per­spec­tives, the in­ter­ac­tion be­tween dis­tor­tions and de­vel­op­ment.

3. How Are Fa­vor­able Dis­tor­tions Pos­si­ble?

This pa­per mainly ad­dresses dis­tor­tions in de­vel­op­ing coun­tries, which usu­ally ex­pe­ri­ence two kinds

3 of mu­tu­ally re­in­forc­ing dis­tor­tions—“en­doge­nous dis­tor­tions” and “pol­icy-im­posed dis­tor­tions.” In the pa­per, we pay more at­ten­tion to the lat­ter, which is be­lieved to be able to con­trib­ute to the eco­nomic take­off and catch up.

In light of neo­clas­si­cal eco­nom­ics, catch up of de­vel­op­ing coun­tries is a nat­u­ral process thanks to

4

“con­ver­gence.” In prac­tice, how­ever, con­ver­gence is of­ten con­di­tional and, in­deed, the gap be­tween de­vel­oped and de­vel­op­ing coun­tries is in many cases even get­ting big­ger. In some East Asian economies in­clud­ing Ja­pan, South Korea, Sin­ga­pore and China’s Tai­wan, gov­ern­ment in­ter­ven­tions are some­times con­sid­ered to be a con­tribut­ing fac­tor to their im­pres­sive growth per­for­mance (World Bank 1993). In ad­di­tion to these East Asian economies, strong gov­ern­ment in­ter­ven­tions (such as tar­iff bar­ri­ers, ex­ploita­tion of colonies and over­seas mar­kets, and in­fant in­dus­try pro­tec­tion) can be found in many

5

Western coun­tries dur­ing their early in­dus­tri­al­iza­tion process. Although these gov­ern­ment in­ter­ven­tions are of­ten dis­tort­ing in the light of neo­clas­si­cal eco­nom­ics, their pos­i­tive im­pact on eco­nomic de­vel­op­ment may gen­er­ally sur­pass their neg­a­tive im­pact, and thus they can be con­sid­ered as “fa­vor­able” dis­tor­tions.

Hence, how are the fa­vor­able dis­tor­tions pos­si­ble? What are their pre­con­di­tions? What are the logic and mech­a­nism un­der­ly­ing them? In what fol­lows, we pro­vide four ar­gu­ments on those ques­tions— ad­van­tage of back­ward­ness, sec­ond- best prin­ci­ple, co­or­di­na­tion fail­ure and po­lit­i­cal econ­omy per­spec­tive.

3.1 Ad­van­tage of Back­ward­ness

Ger­schenkron (1951) firstly dealt with “eco­nomic back­ward­ness” and pointed out that a coun­try, such as the Soviet Union back­ward rel­a­tive to Bri­tain when it em­barked on in­dus­tri­al­iza­tion, did not go through the same stages. His the­ory pre­dicts that the more “eco­nom­i­cally back­ward” a coun­try is, the more we will see, for ex­am­ple, more rapid rates of in­dus­trial growth and a more ac­tive role for the gov­ern­ment and large banks in sup­ply­ing cap­i­tal and en­trepreneur­ship.

A de­vel­op­ing econ­omy can also be at an ad­van­tage be­cause of its rel­a­tive back­ward­ness— in that it can bor­row tech­nolo­gies, busi­ness mod­els and mar­ket­ing pro­ce­dures from more ad­vanced economies; and in that imi­ta­tion may be eas­ier and faster than orig­i­nal in­no­va­tion on which the lead­ing economies have to rely. How­ever, when the ad­van­tage of back­ward­ness grad­u­ally nar­rows and the un­cer­tainty of fron­tier in­no­va­tion in­creases, the gov­ern­ment will find it more dif­fi­cult to col­lect enough in­for­ma­tion and make cor­rect de­ci­sions. The “gov­ern­ment-pick­ing-the-win­ner” model turns out to be a fail­ure. That is why, as economies move to higher in­comes and rely more on in­no­va­tion as the en­gine of growth, the gov­ern­ment in­ter­ven­tion as a ma­jor form of dis­tor­tions will am­plify its ad­verse ef­fects on the econ­omy. There­fore mit­i­gat­ing dis­tor­tions and let­ting the mar­ket play a de­ci­sive role in re­source al­lo­ca­tion are the key to sus­tain­able de­vel­op­ment and there­fore also the key to avoid­ing mid­dle-in­come trap.

3.2 Sec­ond-Best Prin­ci­ple

An­other fac­tor of in­ter­est is the “sec­ond-best prin­ci­ple.” The sec­ond-best prin­ci­ple (Lipsey and Lan­caster 1956) im­plies that if it is not fea­si­ble to re­move a par­tic­u­lar mar­ket dis­tor­tion, in­tro­duc­ing a sec­ond (or more) mar­ket dis­tor­tion(s) may par­tially coun­ter­act the first, and lead to a more ef­fi­cient out­come. In this re­gard, the ar­gu­ment put for­ward by Qian (2016) is of di­rect rel­e­vance to the topic of this pa­per. As he pointed out, some fa­vor­able dis­tor­tions are pos­si­ble given the “sec­ond-best prin­ci­ple.” Con­trary to the “first-best prin­ci­ple,” claim­ing that a dis­tor­tion must lead to a de­cline in ef­fi­ciency, the “sec­ond-best prin­ci­ple” means that in the pres­ence of mul­ti­ple dis­tor­tions, the re­duc­tion of one dis­tor­tion

does not nec­es­sar­ily in­crease ef­fi­ciency, and the ad­di­tion of an­other dis­tor­tion does not nec­es­sar­ily de­crease ef­fi­ciency.

Ar­guably, “rent seek­ing” be­hav­ior, which is of­ten caused by gov­ern­ment reg­u­la­tions, in­dus­trial poli­cies and de­vel­op­ment strate­gies, may im­prove the mar­ke­ti­za­tion process in the pres­ence of in­sti­tu­tional bar­ri­ers. Nev­er­the­less, in the early stage of de­vel­op­ment and in­dus­tri­al­iza­tion, both en­trepreneur­ship and cap­i­tal are scarce re­sources. In this con­text, in­vest­ments, learn­ing and in­no­va­tions may be en­cour­aged by rent seek­ing be­hav­ior—as long as the lat­ter is con­sis­tent with the prin­ci­ple of mar­ket-ori­ented re­form. Since both gov­ern­ment reg­u­la­tions and in­sti­tu­tional bar­ri­ers on the one hand, and rent seek­ing be­hav­ior on the other, can be seen as dis­tor­tions, thus in this sense, a dis­tor­tion can be used to tackle an­other. In re­cent years, this logic has been dis­cussed in the lit­er­a­ture on gov­ern­ment be­havio(rBard­han, 2016).

Ob­vi­ously, China’s re­form pro­ceeds in the pres­ence of many dis­tor­tions, es­pe­cially the in­her­ent “orig­i­nal sin” dis­tor­tions (such as un­der­de­vel­op­ment of mar­ket sys­tem and in­ad­e­quate pro­tec­tion of prop­erty rights), and it needs to be “ad­justed” by dis­tor­tion of other kinds, such as ac­tive gov­ern­ment in­ter­ven­tion. For in­stance, when prop­erty rights are se­cured, the ef­fi­ciency of pri­vate en­ter­prise is gen­er­ally high, and other forms of own­er­ship of en­ter­prises, due mainly to the prin­ci­pal–agent prob­lem, will cause dis­tor­tions. But if prop­erty rights are in­se­cure be­cause of im­per­fect rule of law, then pri­vately owned firms will need to pay to se­cure pro­tec­tion for their prop­erty. There­fore, in this con­text the re­form is likely to choose one dis­tor­tion to deal with an­other, such as the use of lo­cal gov­ern­ment power to pro­tect prop­erty rights against a higher level of gov­ern­ment, which may im­prove ef­fi­ciency. In this sense, some dis­tor­tions may be rea­son­able and nec­es­sary. But de­spite their fa­vor­able ef­fects un­der some cir­cum­stances, dis­tor­tions as tran­si­tional in­sti­tu­tional ar­range­ments in line with the sec­ond-best prin­ci­ple may in­cur high costs, and even be­come sig­nif­i­cant bar­ri­ers to fur­ther deep­en­ing re­forms as an econ­omy en­ters a higher stage of ma­tu­rity.

3.3. Co­or­di­na­tion Fail­ure

Gen­er­ally speak­ing, dif­fer­ent types of gov­er­nance should be con­sis­tent with dif­fer­ent goals of the or­ga­ni­za­tion. At dif­fer­ent stages of de­vel­op­ment, an econ­omy may suf­fer from var­i­ous kinds of co­or­di­na­tion fail­ure, and in this con­text, gov­ern­ment (or the state), mar­ket, com­mu­nity and so forth all can play a role as “co­or­di­na­tor” ( ei­ther com­ple­men­tary or con­tra­dic­tory) in deal­ing with this chal­lenge. Clearly, the ap­pli­ca­bil­ity of the co­or­di­na­tion mech­a­nism or ar­range­ment highly de­pends on the so­cioe­co­nomic en­vi­ron­ments and thus, to avoid a sim­plis­tic and ahis­tor­i­cal per­spec­tive, one should em­pha­size that there is no uni­ver­sal and op­ti­mal model for this.

6

In the­ory, the mar­ket may be an out­stand­ing co­or­di­na­tor of non-co­op­er­a­tive in­ter­ac­tions, in­ef­fi­cient gov­er­nance and per­for­mance in­cen­tives. How­ever, if the resid­ual claim and con­trol are mis­al­lo­cated, and the long-run in­vest­ment de­ci­sions have im­por­tant strate­gic com­ple­men­tar­ity, the mar­ket as a co­or­di­na­tor ceases to be ef­fi­cient. In par­tic­u­lar, the poor tend to be more af­fected by im­per­fec­tions and in­ad­e­quacy of credit and in­sur­ance mar­kets, which dis­cour­ages pro­duc­tive in­vest­ments, in­no­va­tions and hu­man re­source de­vel­op­ment. In this con­text, non-co­op­er­a­tive in­ter­ac­tions tend to be in­ef­fi­cient and there­fore co­or­di­na­tion and se­lec­tive in­cen­tives by the gov­ern­ment are needed to stim­u­late co­op­er­a­tive ac­tions among mar­ket agents.

Mar­ket fail­ure in co­or­di­na­tion pro­vides the ba­sis for gov­ern­ment in­ter­ven­tion. In spite of dif­fer­ent fo­cus, this ar­gu­ment is in the same spirit as the afore­men­tioned “sec­ond best prin­ci­ple.” In some ar­eas and so­cioe­co­nomic en­vi­ron­ments (such as un­der­de­vel­oped mar­kets), the hand of gov­ern­ment is indispensable to deal with mar­ket fail­ure. There­fore, in ad­di­tion to the role of night watch­man and

pro­tec­tor of prop­erty rights, gov­ern­ment may play a mul­ti­ple role as co­or­di­na­tor, guide and stim­u­la­tor, es­pe­cially in the con­text of struc­tural changes and an un­cer­tain de­vel­op­ment path. As a mat­ter of fact, de­vel­op­ment and tran­si­tion coun­tries ap­pear to en­counter more chal­lenges from co­or­di­na­tion fail­ure and thus gov­ern­ment should be associated with proper ca­pac­ity in mo­bi­liz­ing re­sources, mak­ing de­ci­sions, reg­u­lat­ing mar­kets, en­forc­ing laws, and col­lect­ing and treat­ing in­for­ma­tion. Ide­ally, gov­ern­ment should re­main neu­tral re­gard­ing spe­cific in­ter­ests, and be sub­jected to in­sti­tu­tional bal­ance and checks to avoid abuse of gov­ern­ment power. But although gov­ern­ment in­ter­ven­tions aim to tackle mar­ket fail­ure, they may also be sub­jected to malfunction and im­per­fec­tion of their own. For ex­am­ple, although in­dus­trial poli­cies re­gained their pop­u­lar­ity in the wake of the 2008 fi­nan­cial cri­sis, the ad­verse ef­fects of the gov­ern­ment-pick­ing-the-win­ner model (such as rent seek­ing be­hav­ior) have be­come more ev­i­dent. It im­plies that when in­ter­ven­ing in the mar­ket, gov­ern­ment should tackle both co­or­di­na­tion fail­ure and its own de­fi­ciency. Ob­vi­ously, in prac­tice it is ex­tremely hard to do so, es­pe­cially to suc­ceed in strik­ing a bal­ance be­tween mar­ket mech­a­nism and pol­icy dis­tor­tions. That just shows that fa­vor­able dis­tor­tions, which are not al­ways easy to see, are in­deed only pos­si­ble un­der cer­tain strict con­di­tions.

3.4 Per­spec­tive of Po­lit­i­cal Econ­omy

Dis­tor­tion is mostly de­fined from an eco­nom­ics per­spec­tive and thus ex­am­ined with ref­er­ence to mar­ket equi­lib­rium. If tak­ing the some­what broader per­spec­tive of po­lit­i­cal econ­omy, which may take ac­count of as­pects be­yond eco­nom­ics such as na­tional se­cu­rity, geopo­lit­i­cal or ide­o­log­i­cal fac­tors and so forth, some dis­tor­tions could be taken as nor­mal or fa­vor­able. For in­stance, the suc­cess of the “two bombs, one satel­lite” pro­gram in China be­fore mar­ket re­forms and open­ing up is highly praised in the Bei­jing Con­sen­sus, but ap­par­ently con­flicts with the nar­rowly de­fined prin­ci­ple of com­par­a­tive ad­van­tage or cost–ben­e­fit anal­y­sis. An­other ex­am­ple is to sac­ri­fice the in­ter­ests of farm­ers to give pri­or­ity to the de­vel­op­ment of heavy in­dus­try (also from the ex­pe­ri­ence of China). These ex­am­ples

7 in­di­cate that whether a pol­icy is fa­vor­able or not de­pends on which bench­mark or po­si­tion is cho­sen.

4. In­ter­na­tional Ev­i­dence 4.1. De­scrip­tive Anal­y­sis

Our anal­y­sis of in­ter­na­tional ex­pe­ri­ences is based on a sam­ple of 50 coun­tries of dif­fer­ent in­come lev­els. It in­cludes ma­jor economies in var­i­ous con­ti­nents and at dif­fer­ent stages of de­vel­op­ment. No­tably, given the fact that the ag­gre­ga­tion of the 50 coun­tries un­der con­sid­er­a­tion ac­counts for 80% and 90% of world pop­u­la­tion and GDP, re­spec­tively, this sam­ple is highly rep­re­sen­ta­tive. More­over, this rep­re­sen­ta­tive sam­ple will also help to tackle the po­ten­tial bias due to ex­tremely small economies.

Re­gard­ing the mea­sure of eco­nomic dis­tor­tions, we rely on the In­dex of Eco­nomic Free­dom (IEF for short; over­all in­dex), which is jointly pub­lished by the Her­itage Foun­da­tion and the Wall Street Jour­nal. As a com­monly used in­di­ca­tor of eco­nomic free­dom, or, ac­cord­ingly, an op­po­site in­di­ca­tor of dis­tor­tions, the over­all IEF is com­posed of 10 sub-in­dexes, in­clud­ing “prop­erty rights,” “free­dom from cor­rup­tion,” “fis­cal free­dom,” “gov­ern­ment spend­ing,” “busi­ness free­dom,” “la­bor free­dom,” “mone­tary free­dom,” “trade free­dom,” “in­vest­ment free­dom” and “fi­nan­cial free­dom.” All the sub-in­dexes are

8 val­ued from 0 (least free) to 100 (most free). Thanks to its broad cov­er­age of a num­ber of di­men­sions of mar­ket func­tion­ing and eco­nomic ac­tiv­i­ties, it is be­lieved that the IEF can serve as a good in­di­ca­tor for the multi-faceted dis­tor­tions we dis­cussed above.

our con­jec­ture that mar­ket dis­tor­tions will con­trib­ute to TFP growth in the early stages of de­vel­op­ment, and, as the econ­omy ex­pands this pos­i­tive im­pact of dis­tor­tions, will fade away and even change its sign.

5. Ev­i­dence from China

This sec­tion turns to the ex­pe­ri­ences of China since the re­form and open­ing up in 1978. Specif­i­cally, with the help of panel anal­y­sis, we ex­plore the re­la­tion­ship be­tween China’s pol­icy dis­tor­tions, mea­sured by the Mar­ke­ti­za­tion In­dex, and the prov­ince-level (or equiv­a­lent re­gion) TFP growth.

5.1. Dis­tor­tions Mea­sured by Mar­ke­ti­za­tion In­dex

As a com­monly used in­di­ca­tor of mar­ket de­vel­op­ment or dis­tor­tions, the pro­vin­cial Mar­ke­ti­za­tion In­dex (MI) is com­piled by the Na­tional Eco­nomic Re­search In­sti­tute (NERI). It is a com­pos­ite mea­sure of China’s mar­ke­ti­za­tion process in five di­men­sions: gov­ern­ment–mar­ket re­la­tions, de­vel­op­ment of non-state econ­omy, de­vel­op­ment of prod­uct mar­ket, de­vel­op­ment of fac­tors mar­ket, and de­vel­op­ment of in­ter­me­di­ary and le­gal en­vi­ron­ment. The MI and all of its com­po­nents are mea­sured on a scale from 0–10. A prov­ince scores a higher MI when it stays in a lead­ing po­si­tion com­pared with other prov­inces in its progress to­wards mar­ket econ­omy and suf­fered fewer dis­tor­tions com­pared with lower MI prov­inces. Here we use some sta­tis­ti­cal charts to ex­plain the ba­sic facts and char­ac­ter­is­tics of China’s mar­ke­ti­za­tion process in 1997–2014.

First, China’s mar­ke­ti­za­tion in gen­eral moves for­ward, while its speed varies be­tween dif­fer­ent pe­ri­ods. China’s MI con­tin­ued to rise dur­ing 1997–2014 (the only ex­cep­tion is 2010, slightly down by 1.36%), in­di­cat­ing that China’s mar­ket–ori­ented re­form has been ef­fec­tively pur­sued. Over­all, the growth rate of MI varies widely be­tween dif­fer­ent pe­ri­ods. (1) From 1999–2004, the MI in­creased rapidly, prob­a­bly due to China’s ac­ces­sion to the World Trade Or­ga­ni­za­tion (WTO). (2) From 2005– 2010, the MI con­tin­ued to rise, but the growth rate grad­u­ally slowed. (3) From 2011–2014, growth of the MI picked up and the av­er­age an­nual grow rate was about 5%.

As il­lus­trated in Fig­ure 4, the vari­a­tion of the MI be­tween prov­inces ex­panded af­ter 2008, and its dis­tri­bu­tion is roughly con­sis­tent with the level of re­gional eco­nomic de­vel­op­ment; the eastern re­gion as a whole had the high­est MI, fol­lowed by the cen­tral re­gion, and then the western re­gion, which had the low­est score.

11

Fur­ther­more, each MI di­men­sion fol­lows a dif­fer­ent trend. Most di­men­sions of the MI showed an up­ward trend in most pe­ri­ods (Fig­ures 6). The di­men­sion with a clear down­ward trend is Gov­ern­ment– Mar­ket Re­la­tions. In 2006–2013, the Gov­ern­ment–Mar­ket re­la­tions di­men­sion fell from 7.18 to 5.70, and rose slightly to 6.02 in 2014. The Gov­ern­ment–Mar­ket re­la­tions di­men­sion con­sists of three in­di­ca­tors: the pro­por­tion of re­sources al­lo­cated by the mar­ket, the gov­ern­ment in­ter­ven­tion in the en­ter­prise and the gov­ern­ment size. The three in­di­ca­tors fell by 1.33, 0.77 and 1.53 points, re­spec­tively, in­di­cat­ing that since the fi­nan­cial cri­sis of 2008, the Chi­nese gov­ern­ment has strength­ened its power in the re­sources al­lo­ca­tion and ex­panded its size.

5.2. Pro­vin­cial TFP Growth in China

As sug­gested in some re­search, com­pared with the de­vel­oped coun­tries, the con­tri­bu­tion of China’s TFP to eco­nomic growth has yet to be fur­ther im­proved. For in­stance, Brandt and Zhu (2010) de­com­posed China’s eco­nomic growth since the re­form and open­ing-up pol­icy in 1978, and found the con­tri­bu­tion of TFP to eco­nomic growth in China was 62% in 1978–1988, 48% in 1988–1998 and 47%

in 1998–2008, with a grad­u­ally de­clin­ing ten­dency. In this sec­tion, we count TFP for prov­inces from 1978–2015 by us­ing the Solow growth cal­cu­la­tion method. Since the Solow model is the most widely used method for cal­cu­lat­ing TFP, our im­prove­ment is mainly about the es­ti­ma­tion of pro­duc­tion fac­tor in­puts. Based on TFP growth, we can iden­tify the main driv­ing force of China’s eco­nomic growth and judge whether the Chi­nese econ­omy has en­tered the ef­fi­ciency-driven stage.

5.2.1 Equa­tion of TFP cal­cu­la­tion

The pro­duc­tion func­tion is char­ac­ter­ized as a Cobb–Dou­glas pro­duc­tion func­tion:

(3)

A is TFP, α is the cap­i­tal out­put elas­tic­ity, (1–α) is the hu­man cap­i­tal out­put elas­tic­ity. De­riv­ing the de­riv­a­tive of time t, equa­tion (1) changes to

(4) Fur­ther­more, we use the Ho­drick– Prescott ( HP) fil­ter­ing method to re­move the sto­chas­tic per­tur­ba­tion fac­tor of , and fi­nally es­ti­mate the growth rate of the TFP .

5.2.2 In­put ac­count­ing

(1) Phys­i­cal cap­i­tal

The widely used method for es­ti­mat­ing the phys­i­cal cap­i­tal stock is the per­pet­ual in­ven­tory method ini­ti­ated by Gold­smith (1951): . To use this method, four pa­ram­e­ters need to be de­ter­mined: the cap­i­tal stock in the base pe­riod K, in­vest­ment in ev­ery pe­riod I, cap­i­tal price in­dex i and de­pre­ci­a­tion rate δ. Firstly, learn­ing from Young (2000), we set 1952 as the base pe­riod, and mul­ti­ply the in­vest­ment in fixed as­sets in 1952 by 10 to ob­tain the cap­i­tal stock for that year. Se­condly, we use the cap­i­tal for­ma­tion in ev­ery year as the in­vest­ment in­cre­ment. Thirdly, for the cap­i­tal price in­dex, in 1952–1990, we use the re­tail price in­dex as a proxy for it; and in 1991–2015, we use the price de­fla­tor of fixed as­set in­vest­ment to de­flate the value of cap­i­tal stock. Fourthly, the de­pre­ci­a­tion rate from 1952 to 1977 was cal­cu­lated to be 5%, and it in­creased by 0.1% from 1978. This ap­proach is adopted from Wang et al. (2009).

(2) Hu­man cap­i­tal

Ac­cord­ing to the Solow model, the hu­man cap­i­tal could be de­fined as the prod­uct of the num­ber of work­ers (L) and the av­er­age num­ber of years of ed­u­ca­tion (H) of la­bor. The num­ber of la­bor (L) could be ob­tained from the China Sta­tis­ti­cal Year­book. The av­er­age ed­u­ca­tion year (H) could be cal­cu­lated based on the cen­sus data of 1982, 1990, 2000, 2005 and 2010.

(3) Cap­i­tal out­put elas­tic­ity

Chen et al. (1988), Chow and Li (2002) and Lu and Cai (2016) used the ag­gre­gate pro­duc­tion func­tion to es­ti­mate the cap­i­tal out­put elas­tic­ity. Bai and Zhang (2015) ar­gued that the share of cap­i­tal in­come from the ag­gre­gate pro­duc­tion func­tion is con­stant, but this is in­con­sis­tent with China’s real­ity. There­fore, we use the in­come ap­proach to cal­cu­late the cap­i­tal in­come share of each prov­ince as a cap­i­tal out­put elas­tic­ity. The cal­cu­la­tion for­mula is as fol­lows:

(4) Statis­tics of TFP

Fol­low­ing the above method, we es­ti­mate China’s pro­vin­cial TFP growth. The statis­tics for the vari­ables are as fol­lows:

Fig­ure 7 presents the TFP in the three re­gions13 of China. In 1978–2015, TFP steadily in­creased in each re­gion but at a dif­fer­ent pace: TFP in the eastern re­gion is sig­nif­i­cantly faster than in the cen­tral and western re­gions. The gap of TFP be­tween cen­tral and western grad­u­ally in­creased.

TFP growth of each re­gion is shown in Fig­ure 8. In most years from 1978 to 2015, China’s eastern, cen­tral and western re­gions saw pos­i­tive TFP growth, but the growth rate fluc­tu­ated greatly in dif­fer­ent years. In the 1980s and 1990s, TFP growth showed a ten­dency to firstly in­crease and then de­crease. From 2000, the TFP growth trend sta­bi­lized, and af­ter 2010 it sig­nif­i­cantly de­clined.

5.3. Em­pir­i­cal Model

In this sec­tion, we estab­lish a panel re­gres­sion model to fur­ther in­ves­ti­gate the long- term 14 quan­ti­ta­tive con­tri­bu­tions of mar­ke­ti­za­tion to to­tal pro­vin­cial TFP.

We con­struct the fol­low­ing em­pir­i­cal re­gres­sion model.

(6) The sub­scripts i and t rep­re­sent the prov­inces and years. The ex­plana­tory vari­able TFP of the

re­gres­sion model is set as the TFP growth rate of the prov­ince i in year t. The core ex­plana­tory vari­able MIi, t is the Mar­ke­ti­za­tion In­dex ( MI) of prov­ince i in year t. The fixed ef­fect elim­i­nates all the vari­a­tions among prov­inces that do not change over time, while the time-fixed ef­fect re­moves all timelevel dis­tur­bances that do not vary with the prov­ince. rep­re­sents the resid­ual term of the re­gres­sion model.

We fo­cus on the es­ti­mated co­ef­fi­cients of the in­de­pen­dent vari­able MIi, t. The eco­nomic con­no­ta­tion of the co­ef­fi­cient is in­ter­preted as the av­er­age ef­fect of the MI on the de­pen­dent vari­able, that is, the in­flu­ence of the MI on the TFP growth rate.

Other con­trol vari­ables of the re­gres­sion in­clude: (1) GDP per capita, which is used to con­trol the lev­els of eco­nomic de­vel­op­ment. (2) Gov­ern­ment size, which is de­fined as gov­ern­ment ex­pen­di­ture as a pro­por­tion of GDP. (3) Open­ness, which is de­fined as the ex­port share of GDP. (4) In­vest­ment rate, which is de­fined as the share of cap­i­tal for­ma­tion in GDP.

Re­gard­ing sam­ple group­ing, most of the prov­inces in China were in the up­per-mid­dle-in­come group and the lower-mid­dle-in­come group from 1997 to 2014.15 There­fore, the sub­sam­ple re­gres­sions are only for the lower-mid­dle-in­come and the up­per-mid­dle-in­come group. The re­main­ing em­pir­i­cal strate­gies are con­sis­tent with the in­ter­na­tional ev­i­dence sec­tion.

5.4 Re­gres­sion Re­sults

Ta­ble 4 demon­strates the re­gres­sion re­sults of equa­tion (6). Col­umns (1)–(4) of Ta­ble 4 show the re­gres­sion re­sults for fixed ef­fect us­ing the full sam­ple, and col­umns (5)–(6) show the sub­sam­ple re­gres­sion re­sults. In col­umns (1)–(3), the MI has no sig­nif­i­cant ef­fect on TFP growth. In col­umn (4), both of MI's pri­mary and se­condary terms are sig­nif­i­cant, which con­firms the non­lin­ear re­la­tion­ship be­tween dis­tor­tion and eco­nomic growth: that is, the im­pact of MI on TFP growth first de­clines and then in­creases. The in­flec­tion point is 8.38, which cor­re­sponds to the mar­ket-wide level of the whole coun­try in 2013–2014. Col­umns (5) and (6) show that the MI has a sig­nif­i­cant ef­fect on the growth rate of TFP in the up­per-mid­dle-in­come group, but not in the lower-mid­dle-in­come group. For the up­per-mid­dlein­come group, MI in­creases by 1 point and the growth of TFP in­creases by 0.4% on av­er­age.

For other con­trol vari­ables, GDP per capita ( GDPPC) has a neg­a­tive ef­fect on TFP growth, in­di­cat­ing that prov­inces with a high de­vel­op­ment level might lose the late-mover ad­van­tage and have more dif­fi­culty in pro­mot­ing TFP growth. Trade open­ness ( TradeOpen) has sig­nif­i­cantly pos­i­tive co­ef­fi­cients, which is in line with the in­ter­na­tional ev­i­dence. The in­vest­ment rate ( In­vRa­tio) has neg­a­tive ef­fects on TFP growth in ev­ery col­umn, and it has a greater neg­a­tive ef­fect on the lower-mid­dle-in­come group. This in­di­cates that there are large num­bers of in­ef­fi­cient in­vest­ments in China, and the growth driven by low-ef­fi­ciency in­vest­ments is not con­ducive to TFP growth, con­firm­ing the views of Woo (1998), and Bai and Zhang (2015).

The ev­i­dence from China is gen­er­ally con­sis­tent with in­ter­na­tional ex­pe­ri­ence. But it also has its own char­ac­ter­is­tics. Both the ev­i­dence from China and in­ter­na­tional ex­pe­ri­ence con­firm the non- lin­ear re­la­tion­ship of dis­tor­tion for TFP growth. With the de­vel­op­ment of the econ­omy, the pro­mo­tion of mar­ke­ti­za­tion on the growth of TFP will first de­cline and then rise, that is, the ef­fect of dis­tor­tion on the growth of TFP will rise first and then de­cline. In the in­ter­na­tional ex­pe­ri­ence, how­ever, the sup­pres­sion ef­fect of dis­tor­tions oc­curs in the high-in­come group; while in China, the sup­pres­sion ef­fect of dis­tor­tions oc­curs in the up­per-in­come group. Since China has too few sam­ples in high-in­come groups, we can­not ob­tain the re­gres­sion re­sults of high-in­come groups. How­ever, com­bin­ing in­ter­na­tional ex­pe­ri­ence with the Chi­nese ex­pe­ri­ence, we can see that as more prov­inces en­ter the up­per-mid­dle-in­come and high-in­come stage, the neg­a­tive ef­fects of dis­tor­tions on TFP growth will

fur­ther man­i­fest them­selves.

5.5 Ro­bust­ness Anal­y­sis

To check the ro­bust­ness of the pre­vi­ous find­ings, we use the sub-in­dexes of MI as al­ter­na­tive mea­sures for pol­icy dis­tor­tions. To in­ves­ti­gate the non-lin­ear re­la­tion­ship be­tween dis­tor­tions and TFP growth, we also di­vide the sam­ple pe­riod into two sub-sam­ples: lower-mid­dle-in­come group and high­er­mid­dle-in­come group. Other con­trol vari­ables re­main un­changed. The re­sults are shown in Ta­ble 5.16

The Gov­ern­ment–Mar­ket Re­la­tions (MI1) have a sig­nif­i­cant neg­a­tive ef­fect on TFP growth in the lower-mid­dle-in­come group, and a sig­nif­i­cantly pos­i­tive ef­fect in the up­per-mid­dle-in­come group.

The De­vel­op­ment of Non-State Eco­nomic (MI2) pro­moted TFP growth in the lower-mid­dle-in­come group, but not in the up­per-mid­dle-in­come group. The rea­son may be that af­ter un­der­go­ing a largescale re­form of state-owned en­ter­prises, the Chi­nese econ­omy has formed a ver­ti­cal struc­ture, of which up­stream in­dus­tries (such as en­ergy, fi­nance and telecom­mu­ni­ca­tions) are still oc­cu­pied by state-owned en­ter­prises, and most down­stream in­dus­tries (such as con­sumer goods, man­u­fac­tur­ing, and con­sumer ser­vices such as ho­tel and en­ter­tain­ment in­dus­tries) have been dom­i­nated by pri­vate econ­omy (Li et al. 2014). With China's ac­ces­sion to the WTO, both down­stream in­dus­try and up­stream in­dus­try have greatly im­proved. There­fore, in the up­per-mid­dle-in­come group, the state-owned econ­omy and the non­state-owned econ­omy are equally im­por­tant for im­prov­ing TFP growth. The rel­a­tive share of the non­state-owned econ­omy in the over­all econ­omy has no sig­nif­i­cant ef­fect on TFP growth.

The De­vel­op­ment of Fac­tors Mar­ket (MI4) has no sig­nif­i­cant ef­fect on TFP growth in the low­er­mid­dle-in­come group, and a sig­nif­i­cant pos­i­tive ef­fect on TFP in the up­per-mid­dle-in­come group. The re­port of the 19th Na­tional Congress of the CPC pointed out that eco­nomic sys­tem re­form must fo­cus on im­prov­ing the prop­erty rights sys­tem and the mar­ket-ori­ented al­lo­ca­tion of fac­tors. Fur­ther im­prov­ing the al­lo­ca­tion ef­fi­ciency of fac­tor mar­ket is an im­por­tant guar­an­tee for the sus­tain­able de­vel­op­ment of the econ­omy.

Fi­nally, the De­vel­op­ment of In­ter­me­di­ary and Le­gal En­vi­ron­ment (MI5) has a sig­nif­i­cant pos­i­tive im­pact on TFP growth in both the lower-mid­dle-in­come group and the up­per-mid­dle-in­come group. The re­gres­sion re­sults for other vari­ables are con­sis­tent with those in Ta­ble 4.

6. Con­clud­ing Re­marks and Pol­icy Im­pli­ca­tions

Our em­pir­i­cal re­sults based on in­ter­na­tional ex­pe­ri­ences show that dis­tor­tions can hin­der the en­gine of long-term growth, mea­sured by the TFP, in high-in­come economies. By con­trast, dis­tor­tions may help mid­dle-in­come economies to achieve faster TFP growth un­der cer­tain con­di­tions.

Em­pir­i­cal study of China’s prov­inces gen­er­ally shows that mar­ke­ti­za­tion sig­nif­i­cantly pro­moted the growth of the coun­try’s TFP dur­ing the pe­riod 1997–2014. That is to say, dis­tor­tions have sig­nif­i­cant neg­a­tive ef­fects on the coun­try’s TFP growth dur­ing this pe­riod. How­ever, in the early stage of de­vel­op­ment (1978–1991), dis­tor­tions played a sig­nif­i­cantly pos­i­tive role in the TFP growth.

Over­all, de­spite some slight dif­fer­ences, the main find­ings from in­ter­na­tional and Chi­nese ex­pe­ri­ence seem ba­si­cally con­sis­tent, so that we can con­clude that in the early stage of de­vel­op­ment, ap­pro­pri­ate dis­tor­tions can help break through the poverty trap. But as the econ­omy climbs to a higher in­come level, the ad­verse ef­fects of dis­tor­tions be­come more and more sig­nif­i­cant and even­tu­ally hin­der sus­tain­able growth.

Our find­ings have im­por­tant im­pli­ca­tions for China’s pol­icy mak­ing. First, the tran­si­tion from mid­dle-in­come to high-in­come econ­omy is also a process of mit­i­gat­ing mar­ket dis­tor­tions. Although the lat­ter might have con­trib­uted to the coun­try’s eco­nomic take­off and growth catchup in the past, se­ri­ous prob­lems due to dis­tort­ing mar­ket sys­tems have been cre­ated and be­come sub­stan­tial im­ped­i­ments to re­form and de­vel­op­ment in the fu­ture. In view of this, to avoid the so-called mid­dle-in­come trap and to achieve sus­tain­able de­vel­op­ment, the key is to pro­mote mar­ket-ori­ented re­forms and, in par­tic­u­lar,

17 re­duce in­ap­pro­pri­ate gov­ern­ment in­ter­ven­tions and in­sti­tu­tional dis­tor­tions.

Sec­ond, it by no means im­plies that the econ­omy can func­tion well with­out gov­ern­ment. From a

the­o­ret­i­cal stand­point, gov­ern­ment ser­vices and dis­tor­tions dif­fer in many im­por­tant ways. In par­tic­u­lar, an ef­fi­cient gov­ern­ment with ap­pro­pri­ate “state ca­pac­ity” is indispensable in a modern mar­ket-based econ­omy. As a mat­ter of fact, it is not un­com­mon that in ad­vanced economies with ma­ture mar­ket sys­tems, gov­ern­ment also plays a proac­tive role in many so­cio-eco­nomic di­men­sions, es­pe­cially in en­cour­ag­ing in­no­va­tion. In other words, the dis­tor­tions per se can be viewed as the sit­u­a­tion in which gov­ern­ment does not as­sume its role prop­erly. What should be done to let the gov­ern­ment play a “bet­ter” role in a mar­ket econ­omy con­text as stated in the Third Ple­nary Ses­sion of the 18th CPC Cen­tral Com­mit­tee? In our view, in­stead of “re­plac­ing” mar­ket mech­a­nism, gov­ern­ment should play a proac­tive role in the “mar­ket-re­in­forc­ing” process. In other words, var­i­ous gov­ern­ment in­ter­ven­tions should only

be judged by their con­tri­bu­tions to mar­ket-ori­ented re­forms, es­pe­cially to help­ing the mar­ket play a de­ci­sive role in re­source al­lo­ca­tion.

Third, with clear ob­jec­tives of eco­nomic re­forms, “fa­vor­able dis­tor­tions” should be sub­jected to strict con­straints. Although “fa­vor­able dis­tor­tion” is in­deed of­ten taken as a sup­port­ive ar­gu­ment for keep­ing dis­tor­tion and the im­por­tant role of gov­ern­ment in the econ­omy, the re­la­tion­ship be­tween dis­tor­tion and eco­nomic per­for­mance is com­plex, es­pe­cially non­lin­ear. In par­tic­u­lar, fa­vor­able dis­tor­tion sub­stan­tially de­pends on var­i­ous so­cioe­co­nomic con­di­tions re­lated to stage of de­vel­op­ment. If these con­di­tions are not met, dis­tor­tions will ham­per de­vel­op­ment. For in­stance, the ad­van­tage of back­ward­ness, sec­ond-best prin­ci­ple, and co­or­di­na­tion fail­ure all have their the­o­ret­i­cal pre­con­di­tions, and, gen­er­ally speak­ing, are ap­pro­pri­ate for de­vel­op­ing coun­tries which have im­ma­ture mar­ket sys­tems, se­ri­ous struc­tural prob­lems and chal­lenges re­gard­ing “take­off” and “tran­si­tion”. Since those coun­tries in­clud­ing China en­ter into a higher level of de­vel­op­ment with an im­proved in­sti­tu­tional en­vi­ron­ment, the so-called fa­vor­able dis­tor­tions will be­come less plau­si­ble. For this rea­son, China’s pol­icy mak­ers should take a clear stand to re­duce dis­tor­tions, and let the mar­ket play a de­ci­sive role in re­source al­lo­ca­tion. Oth­er­wise, in the name of “growth catch-up,” the pol­icy-im­posed dis­tor­tion will oc­cur fre­quently and the di­rec­tion of mar­ket-ori­ented re­form will be­come blurred and swing. Mit­i­gat­ing un­fa­vor­able dis­tor­tions is largely a process of ex­plor­ing the fa­vor­able bor­der­line be­tween gov­ern­ment and mar­ket, which con­sti­tutes a ma­jor chal­lenge for all economies.

TFP = to­tal fac­tor pro­duc­tiv­ity.Notes: 1. Stan­dard er­rors in paren­the­ses, with ***, **, * de­not­ing 1%, 5%, and 10% level of sig­nif­i­cance, re­spec­tively. 2. All re­gres­sions above in­clude a con­stant.Source: The au­thors. Ta­ble 5: TFP Growth and the State-Owned Econ­omy (1997–2014)

Newspapers in English

Newspapers from China

© PressReader. All rights reserved.