doesfinancialdevelopmenthamperorimprovetheresource curse

Research ArticleDoes Financial Development Hamper or Improve the ResourceCurse Analysis Based on the Panel Threshold Effect Model

ZhiQiang Sun 12 and ZeXiang Cai3

1Jinshen College Nanjing Audit University Jiangsu Nanjing China2College of Economics and Management Nanjing University of Aeronautics and Astronautics Jiangsu Nanjing China3School of Finance Nanjing Audit University Jiangsu Nanjing China

Correspondence should be addressed to ZhiQiang Sun szqnuaaeducn

Received 31 March 2020 Revised 14 May 2020 Accepted 17 June 2020 Published 15 July 2020

Academic Editor Kauko Leiviska

Copyright copy 2020 ZhiQiang Sun and ZeXiang Cai is is an open access article distributed under the Creative CommonsAttribution License which permits unrestricted use distribution and reproduction in anymedium provided the original work isproperly cited

is paper analyzes the ldquoresource curserdquo and ldquofinancial threshold effectrdquo that may exist in China and then uses the data from 30provinces from 2004 to 2018 as research samples We used linear regression and nondynamic panel threshold models to analyzethe financial threshold effects of the ldquoresource curserdquo hypothesis and the ldquoresource curserdquo phenomenon At the same time wedivided the level of financial development to verify the robustness of the research conclusions in this paper e study found thefollowing (1)ere is a certain correlation between the abundance of resources and economic growthWhether this can be seen asa ldquocurserdquo or a ldquoblessingrdquo of resources is significantly related to the degree of financial development (2) Whether financialdevelopment can alleviate the ldquoresource curserdquo depends on the degree of financial development In the extremely scarce stage offinancial resources the resource endowment effect is obvious and the level of economic development in resource-based regionswill be higher than in other regions when the level of financial development is low (financial resources are not scarce and have notreached a reasonable level) the phenomenon of the ldquoresource curserdquo appears when the level of financial development is highlydeveloped economic development benefits more from financial development and the effects of resource endowment declineOnly when financial development is at a reasonable level can resource endowments effectively raise the level ofeconomic development

1 Introduction

Resource-oriented traditional economic growth theoryholds that resource endowment is necessary for economicgrowth Looking back on the course of modern economicdevelopment Britain the first county to experience an in-dustrial revolution had abundant resources such as coal andore the earliest areas of industrialization in Germany alsoappeared in the resource-intensive industrial area of Ruhrthe fact that the United States thrived in the late 19th andearly 20th centuries is also closely related to wealthy resourceextraction and production [1] However subsequent de-velopment practices have run counter to this and countriesand regions rich in resources often perform poorly Whetherit is Spain Russia or oil-rich countries such as Nigeria and

Venezuela these countries have encountered developmentobstacles to varying degrees On the contrary SwitzerlandJapan Hong Kong Taiwan South Korea Singapore andother countries in the same period of time under thecondition of relatively scarce resources are developing at ahigher speed After the 1950s there was a lot of discussionabout the negative correlation between natural resourcesand economic growth Paypyrakis and Gerlagh [2] verifiedthat the ldquoresource curse phenomenon exists within thecountryrdquo and therefore confirmed that the resource cursephenomenon exists Auty [3] first proposed the ldquoresourcecurserdquo proposition and summarized the negative correlationbetween resource abundance and economic growth as theldquoresource curserdquo effect and pointed out the developmentdifficulties of resource-based economies

HindawiMathematical Problems in EngineeringVolume 2020 Article ID 4365205 10 pageshttpsdoiorg10115520204365205

Abundant natural resources should be the source ofmomentum for economic growth but the ldquotrapsrdquo in eco-nomic development brought about by them have arousedmuch interest especially regarding their theoretical mech-anisms [4] With the increasing amount of research on thetransmission mechanism there are fewer studies focused onsolving the ldquoresource curserdquo Why do resources cause de-velopment traps Exploring the cause of the ldquoresource curserdquoand overcoming resource problems have become an im-portant issue for many resource-abundant countries [5]

China is a big resource-rich country Many resource-rich provinces and cities have contributed to the countryrsquosconstruction However with the exploitation of resourcesmany resource-based areas have encountered developmentalobstacles e questions of how to solve resource-basedproblems and promote resource-based areas are of greatsignificance in terms of helping its transformation [6]Evidence shows that resource abundance will also have anegative impact on the development of finance but moststudies do not mention the relationship between the ldquofi-nancial development levelrdquo and ldquoresource abundancerdquoWhen sorting through the development context of resource-rich countries we found that not all resource-rich countrieshave fallen into a development trap Compared with thetragic fate of some countries that have been ldquocursed byresourcesrdquo other countries have achieved considerable de-velopment by virtue of their resource advantages Researchby Pendergas et al [7] found that ldquoresource countries thathave gotten rid of the development trap have developed theirfinancial development on the contrary resource countriesthat have suffered from curses often have very low levels offinancial developmentrdquo Is this a coincidence Finance is thecore of the modern economy and an effective financialsystem has the important function of rationally allocatingresources and vigorously promoting economic develop-ment Can it also be applied to natural resources On thebasis of the financial perspective this paper explores how thelevel of financial development affects the resources curse Onthe basis of the panel data from 15 resource-based regions inChina from 2004 to 2018 the Hansen threshold panel modeland the quasi-irrelevant regression estimation (SUR)method were used to empirically analyze whether the fi-nancial development of resource-based regions would affectthe resource curse

e rest of the paper is organized as follows Section 2 is aliterature review which mainly introduces the research ofdomestic and foreign scholars in this field and provides atheoretical reference for the research of this article Section 3presents a detailed description of the method used in ourstudy Section 4 presents and discusses the results from theempirical analysis Section 5 summarizes the conclusionsand puts forward corresponding countermeasures andsuggestions based on the empirical analysis results

2 Literature Review

e introduction of the ldquoresource curserdquo hypothesis hassparked widespread discussion in academia Foreign re-searchers have previously studied the ldquoresource curserdquo

problem Most of the research conclusions support theldquoresource curserdquo hypothesis with the ldquoresource curserdquotransmission mechanism being proposed from differentdimensions studies [8ndash10] Gylfason [11] systematicallyexpounded on how the transmission mechanism of theldquoresource curserdquo affects economic growth and found thatthere are four transmission paths including ldquoDutch diseaserdquorent-seeking government decision-making errors and hu-man capital investment Subsequently a large number ofscholars enlarged their assumptions on the analysis of theldquoresource curserdquo transmission mechanism For examplePapyrakis and Gerlagh [2] found the existence of threeldquoresource curserdquo transmission mechanisms governmentcorruption foreign trade and education investment Wangand Li [12] used recent panel data at the provincial andmunicipal levels and coal development intensity as ameasurement index to test the ldquoresource curserdquo effect ofChinarsquos coal citiese results show that the coal resources ofthe sample coal cities were significantly positively correlatedwith economic growth which leads to the conclusion thatthe ldquoresource curserdquo is not significant at the city level inChina Lu [13] believes that the root cause of a resource-richcountryrsquos long-term low performance in terms of devel-opmental path is not based on the ldquocurserdquo of abundantnatural resources but should be attributed to the institu-tional conditions and the ldquopath dependencerdquo caused bychanges At the formal regulatory level it is difficult forcountries to benefit from an excess of knowledge and skills ina participation structure associated with resource propertyrights which is further constrained by the dilemma ofproduction capacity and operation and management Interms of constraints on the informal level under thestructure of a rent-seeking culture fluctuations in the re-source cycle weaken the systemrsquos reason distort the in-centive structure and induce institutional changes At thelevel of implementation path the unpredictability of weakintertemporal commitments raises the transaction costs ofthe government entering into and fulfilling contracts withdevelopmental enterprises Peretto [14] expounded on thetransmission mechanism of the ldquoresource curserdquo from theperspective of technological innovation and found that therewas a significant negative correlation between the abun-dance of resources and regional technological innovationcapabilities Bjorvatn et al [15] found that rent resources andpolitical resources are a significant factor for the ldquoresourcecurserdquo and thereafter the transmission mechanism of po-litical resources and political systems became the focus ofresearch by experts and scholars [16] However as towhether it is the ldquoresource blessingrdquo or ldquoresource curserdquo theacademic community has always been undecided and hasnever come to reach a consensus Studies by Song and Li [17]and Alexeev and Conrad [18] all challenge the ldquoresourcecurserdquo hypothesis ey analyzed and found that the re-source endowments of Russia Canada Norway and othercountries have a significant positive effect on economicgrowth At the same time they also pointed out that there isno direct evidence that rich natural resources have a de-structive effect on economic growth Haber and Menaldo[19] by excluding regional heterogeneity and time

2 Mathematical Problems in Engineering

invariance found through normative analysis that in manycases resources have a ldquoblessingrdquo effect

Regarding the research of the ldquoresource curserdquo phe-nomenon foreign scholars mostly focus on qualitative an-alyses while domestic scholars prefer quantitative analysesand most of the conclusions confirm the ldquoresource curserdquohypothesis Xu and Han [20] identified the ldquoresource curserdquoproblem in China from the perspective of the ldquoDutch dis-easerdquo effect and the weakening of the system Sun et al [21]studied the relationship between natural resource depen-dence public education investment and human capitalaccumulation ey found that a dependence on naturalresources and human capital accumulation were signifi-cantly negatively correlated e crowding-out effect ofnatural resources on human capital only exists in the centraland western regions of China e government should usethe income of the natural resources sector to increase in-vestment in education to enhance local human capital Jiangand Liu [22] used the ldquoresource curserdquo hypothesis model totest the relationship between energy resource developmentand economic growth in Central Asia using panel data fromfive countries in Central Asia from 2000 to 2017 Studieshave shown that since the beginning of the 21st centurythere has been a significant positive relationship betweenenergy resource development and economic growth inCentral Asia which excludes the hypothesis that the regionrsquoslong-term dependence on energy resource developmentleads to a ldquoresource curserdquo Sun et al [23] believe that thegovernment should adopt the method of transfer payment topromote regionally balanced development Better economicand social policies will help transform wealth from naturalresources to economic growth erefore a ldquoresourceblessingrdquo may occur instead of a ldquoresource curserdquo Equitabledistribution and high-quality education enhance humancapital and thus promote economic growth forcing a changefrom the current resource-driven model to a knowledge-driven model Ma and Cheng [24] found the transformationmechanism of technological progress in the ldquoresource curserdquoand ldquoresource blessingrdquo phenomena According to theirresearch energy-based enterprises enjoy the ldquoresourceblessingrdquo in the early stages of development When theexpansion of the enterprise reaches a certain threshold thephenomenon of the ldquoresource curserdquo appeared immediatelyWan and Wang [25] first studied the issues related to in-troducing social capital into the ldquoresource curserdquo and foundthat the accelerated accumulation of social capital andtechnological innovation can cut off the transmission path ofthe ldquoresource curserdquo and there is a threshold effect betweenthem Once the threshold is crossed the ldquoresource curserdquo istransformed into the ldquoresource blessingrdquo erefore measuressuch as encouraging technological innovation increasing socialcapital cultivation and improving government efficiency breakthe ldquoresource curserdquo curse Yan [26] used western China as aresearch object analyzing the transition mechanism of policythresholds and the ldquoresource curserdquo issue from the perspectiveof spatial value He believes that since the Silk Road EconomicBelt was jointly constructed the ldquoresource curserdquo has sloweddown in western regions and the ldquoresource blessingrdquo effect hasincreased

In summary in the research of the ldquoresource curserdquohypothesis scholars mostly analyze the ldquoresource curserdquofrom the perspective of resource input government func-tions and technological innovation Few scholars pay at-tention to financial development in the ldquoresource curserdquo rolein question Some studies have inspired us for exampleBadeeb and Lean [27] believe that the financial system canbring rental income to natural resources Although theauthors did not perform a specific analysis at the theoreticalor empirical level it can still be used to highlight how thelevel of financial development is an important factor in theldquoresource curserdquo Yue and Zhang [28] explored the rela-tionship between natural resource abundance and economicgrowth at different stages of financial development eirresearch shows that in China there is a significant nonlinearrelationship between resource abundance and economicgrowth and only if the level of financial development isreasonable can the ldquoresource curserdquo be turned into theldquoresource blessingrdquo However they only discussed thetransmission mechanism on the theoretical level and thusan empirical level analysis is still lacking More importantlythere is a lack of discussion on the formal financial market Itis undeniable that China has a ldquoresource curserdquo problem Toexplore whether financial development can effectively solveChinarsquos ldquoresource curserdquo problem it is necessary to explorewhether financial development is an important transmissionmechanism of the ldquoresource curserdquo affecting economicgrowth

3 Method

31 Panel$resholdModel e threshold effect refers to thephenomenon of when one economic parameter reaches acertain value another economic parameter suddenlychanges to other forms of development the critical value ofthis explanatory variable is the threshold value [29]

311 Model Settings Hansen [30] defined the basic form ofthe ldquothreshold regressionrdquo model as

yi θ1primexi + ei qi le c (1)

yi θ2primexi + ei qi gt c (2)

where xi as an explanatory variable is an m-dimensionalcolumn vector and qi is a ldquothreshold variablerdquo Hansen [30]showed that the threshold variable can be either a regressionelement in the explanatory variable xi or an independentthreshold variable and according to the ldquothreshold valuerdquo csamples can be divided into ldquotwo regimesrdquo

When combining the forms of equations (1) and (2) intoa single form first one needs to define a dummy variabledi(c) qi le c1113864 1113865 and let xi(c) xidi(c) erefore equa-tions (1) and (2) can be written as

yi θprimexi + δnprimexi(c) + ei (3)

rough this method of adding dummy variables we cansee that θ θ2 and δn θ2 minus θ1en we can rewrite equation(3) into a matrix form

Mathematical Problems in Engineering 3

Y Xθ + Xcδn + e (4)

e regression parameters in the model are (θ δn c)When c is given θ and δn in equation (4) have a linearrelationship erefore according to the conditional leastsquares estimation method using Xlowastc X Xr1113858 1113859 for re-gression on Y the corresponding residual sum of squaresfunction is as follows

Sn(c) Sn(θ(c) δ(c) c) YprimeY minus YprimeXlowastc Xlowastprimec Xlowastc1113874 1113875

minus1Xlowastprimec Y

(5)

e estimated threshold value is used to minimize Sn(c)It is defined as

1113954c arg mincisinΓn

Sn(c) (6)

where Γn Γ cap q1 qn1113864 1113865 Hansen [30] took each obser-vation in the threshold variable as a possible threshold anddetermined the observation value that suits equation (6) asthe threshold After the threshold value is determined otherparameter values can be determined accordingly

32 Significance Test e purpose of a significance test is totest whether the model estimation parameters of two groupsof samples divided by threshold values are significantlydifferent erefore the null hypothesis in which nothreshold exists isH0 θ1 θ2e LagrangeMultiplier (LM)statistics are constructed as follows

L nS0 minus Sn(1113954c)

Sn(1113954c) (7)

where S0 is the sum of squared residuals under the nullhypothesis Because the LM statistics do not obey thestandard distribution Hansen [31] proposed the idea ofobtaining the asymptotic distribution by ldquoBootstraprdquo andthen obtained the corresponding probability P value knownas the Bootstrap P value e basic idea of this method isthat under the premise of given explanatory variables andthreshold values a simulation generates a set of dependentvariable sequences and makes it obeyN (0 1113954e2) where 1113954e is theresidual of the item in equation (4) For each self-samplingsample a simulated LM statistic can be calculated isprocess was repeated 1000 times and Hansen [31] showedthat the P value estimated by the ldquoBootstrap methodrdquo wasequal to the number of times the LM statistics generated bythe simulation exceeded that of the formula (equation (7))which was expressed as a percentage of the total number ofsimulations e Bootstrap P value here is similar to thecompanion probability P value obtained by ordinary mea-surement methods For example when the Bootstrap P valueis less than 001 it means that the LM test has passed asignificance level of 1 and so on

33 Confidence Interval When it is determined that there isa ldquothreshold effectrdquo for a certain variable the confidenceinterval of its threshold value needs to be further

determined e null hypothesis H0 Test ldquoLikelihood RatioStatisticrdquo can be expressed as

LRn(c) nSn(c) minus Sn(1113954c)

Sn(1113954c) (8)

Hansen [32] believes that when LRn (c)le c(α) minus2 ln(1 minus α) the null hypothesis cannot be rejected (αindicates the significance level) Here c (α) equals 735 at a95 confidence level

e above test process is a test process with only onethreshold In order to determine whether there are twothresholds or more we should check whether there are twothresholds Rejecting L means that at least one thresholdexists We can assume that we have estimated 1113954c1 and thenstart looking for a second threshold 1113954c2 After determiningthat there are two thresholds the third threshold is soughtusing the same method as before until we cannot reject thenull hypothesis

4 Results Analysis

41 Variable Selection and Data Description

411 Sample Selection Regarding the feasibility of thestudy we used data from 30 provinces in China (excludingTibet Autonomous Region and Hong Kong Macao andTaiwan) from 2004 to 2018 as the research sample whichconstitutes a balanced panel of 450 observations e datamainly come from the National Bureau of Statistics thestatistical yearbooks of the provinces the WIND database(WIND) and the Guo Tairsquoan database (CSMAR)

412 Explained Variable

(1) $e Level of Development of the Real Economy (y) Afterthe outbreak of the financial crisis the US Federal Reservefrequently used the term real economy and defined it as apart other than the financial and real estate industries andother domestic scholars also refer to industries other thanfinance and real estate as the real economy in their researche measurement method is to subtract GDP from theoutput value of the financial industry and real estate in-dustry is paper draws on the practice of scholars and alsouses this method to measure the level of real economydevelopment in each region expressed as y

413 Explanatory Variables

(1) Natural Resource Abundance (res) To test the ldquoresourcecurserdquo hypothesis we first estimated the abundance of re-sources in each region Some scholars use the total wages ofemployees in extractive industriestotal wages of regionalemployees the ratio of investment in extractive industryconstruction to GDP and the value of energy industrialoutputtotal industrial output as indicators of the abundanceof natural resources Because of the large regional differencesin energy distribution these can better reflect the naturalresource endowment characteristics of a region erefore


this paper uses energy abundance to indicate the level ofregional natural resource endowment Here we draw onvarious research methods to measure regional natural re-source abundance

res ω1lowast coal + ω2 oil + ω3 nature gas

coal + oil + nature gas (9)

Among them ω1ω2 and ω3 represent the standard coalconversion reference of raw coal crude oil and natural gasrespectively (the reference coefficients of raw coal crude oiland natural gas to standard coal converted by China EnergyStatistical Yearbook 2013 are 07143 14286 and 133) coaloil and natural gas represent proven reserves of coal oil andnatural gas

(2) Physical Capital Investment (inv) is refers to the use ofthe fixed asset investment amountthe level of developmentof the real economy (y) In order to facilitate the subsequentempirical analysis of the relationship between the real estatemarket finance and the real economy we divided the in-dicator inv into the following groups (fixed asset investmentamount minus real estate development investment amount)the level of development of the real economy (y) as inv1 andthe amount of real estate development investmentthe levelof development of the real economy (y) as inv2 inv2 rep-resents the real estate investment in each region

(3) Human Capital Investment (edu) In order to facilitateanalysis and data collection we borrowed from the calcu-lation methods of scholars such as Zhang and Zhang [33]Herein it is measured by the average years of education

(4) Scientific and Technological Innovation (tech) is refersto the three scientific and technological expensesy as anindex to measure scientific and technological innovationwhere the three scientific and technological expenses refer tothe new product trial production fees intermediate trial feesand subsidies for major scientific research projects estab-lished by the state to support the development of scientificand technological undertakings

(5) Degree of Openness (open) In this paper we used the totalvalue of import and export tradethe level of development ofthe real economy (y) [34]

(6) Corruption (cor) Rent-seeking and corruption are one ofthe ways to ldquocurserdquo resources so this article adds a cor-ruption variable cor which is measured by the number ofcrimes (corruption bribery and misconduct) per 10000public officials

414 $reshold Variables e threshold variable to beverified in this article is the financial development (fin) eselection of the variable of finance development was doneconsidering that in China financial institutions have acentral bank as the core and a commercial bank as the mainbody ie a coexisting financial system erefore given thecharacteristics of this bank-led financial system the financial

development variable should be a bank-related indicator Atthe same time we know that corporate activities are closelyrelated to deposits and loans and financial institutions havesignificant differences in deposits and loans erefore wechose ldquoratio of loans to regional GDPrdquo as the index of fi-nancial development variables

42 Model Construction In order to clarify whether fi-nancial development affects the relationship between naturalresources and the real economy this paper uses the Hansenthreshold model for the empirical test e basic idea of thesingle threshold regression in this article is that there is athreshold c for the financial development level fin in themodel For finge c and finlt c the relationship betweennatural resources and the growth of the real economy hasobvious differences e single threshold panel model is setas follows

yit μi + αXit + β1 resit finit lt c( 1113857 + β2 resit finit ge c( 1113857 + εit

(10)

where i represents each region t represents the year theinterpreted variable yit is the growth rate of the realeconomy resit represents the abundance of natural re-sources Xit is the main variable affecting the growth of thereal economy c is an unknown threshold I(bull) is indicatorfunction β1 and β2 denote the influence coefficients of thethreshold variable fin on the growth of the real economyunder the conditions of finitge c and finitlt c μi is the regionalindividual effect α is the parameter to be estimated and εit isa random error term Multiple thresholds may appearduring the research e triple thresholds are used as anexample to explain the multiple thresholds model emodel of the triple threshold panel is set as follows

yit μi + αXit + β1resit finit lt c1( 1113857 + β2resit c1 lt finit le c2( 1113857

+ β3 resit c2 lt finit le c3( 1113857 + β4 resit finit gt c3( 1113857 + εit

(11)

First the threshold value is estimated (by choosing thefin value that minimizes the sum of the squared residuals ofthe model as the threshold value) then the coefficient valuesof other variables are estimated ereafter a significancetest on the threshold value is performed to test whether thethreshold effect exists and after determining the number ofthresholds a confidence interval for c to test the truth of thethreshold effect is constructed

43 Empirical Results Analysis

431 Descriptive Statistical Analysis Table 1 shows thedescriptive statistics of the variables From the perspective ofthe explanatory variables the mean value of y is 1133 thestandard deviation is 0552 and the range is 2697 estandard deviation is the sum of the square of the differencebetween the data and the average e smaller the standarddeviation the more stable the data From the descriptivestatistics the standard deviation is small indicating that theselected data (y) has a good level of stability erefore the


empirical test mainly uses y as the explanatory variable foranalysis In addition it can also be seen that there are largedifferences in the level of regional economic development inChina From the perspective of explanatory variables theaverage of the abundance of natural resources is 3191 thestandard deviation is 3656 and the range is 15782 whichindicates that there are great differences in resource en-dowments between regions in China e average of fi-nancial development is 2554 the standard deviation is0543 and the range is 2988 indicating that there aresignificant differences in financial development betweenregions in China In terms of control variables the standarddeviations of inv edu open cor and tech are all greater than05 indicating that regional differences are significant

432 Empirical Results Analysis e paper examines theimpact of the level of financial development on the rela-tionship between natural resources and the real economyverifies whether there is an impact and conducts an em-pirical analysis of its impact path

(1) Test and Analysis of the ldquoFinancial $reshold effectrdquoHypothesise threshold test results in Table 2 show that theF values of the single threshold test are 33998 and thecorresponding P values are 0001 indicating that the modelhas a threshold effect e F value of the double thresholdmodel test is 8332 and the corresponding P values are0002 indicating that the model may have two or morethresholdse triple threshold test results show that there isa third threshold indicating a triple threshold effect

(2)$reshold Estimation Regression Results According to thethreshold test results there is a triple threshold effect so theequation is established as followsyit μi + αXit + β1 resit finit lt c1( 1113857 + β2 resit c1 lt finit le c2( 1113857

+ β3 resit c2 lt finit le c3( 1113857 + β4 resit finit gt c3( 1113857 + β5 invit

+ β6 edu + β7 techit + β8 corit + β9 openit + εit

(12)

In this paper we use the uncorrelated regression esti-mation (SUR) method to perform the test e results arepresented in Table 3 [35]

From the regression results in Table 3 when the level offinancial development is low (finle 2431) the regressioncoefficient of res is a significant positive number indicatingthat in the lower level stage of financial development the

regions with higher levels of resource abundance have ahigher level of economic development When the level offinancial development crosses the first threshold but doesnot cross the second threshold (2431lt finle 2445) theregression coefficient of res is a significant negative numberwhich indicates that at this stage the abundance of re-sources cannot significantly promote economic develop-ment and the ldquoresource curserdquo phenomenon has occurredWhen the level of financial development is high(2445lt finle 2625) the regression coefficient of res is asignificant positive number indicating that during periods

Table 1 Variable descriptive statistics

Variable Number of samples Mean Standard deviation Minimum value Maximum valuey 450 1133 0552 minus0298 2399res 450 3191 3656 0 15782fin 450 2554 0543 1033 4021inv 450 3887 0502 2763 4882edu 450 1776 0548 0697 3213tech 450 2109 0631 0612 3229cor 450 1032 0598 0591 3218open 450 1663 0661 0339 2212

Table 2 reshold effect test results

Test type Statistics Test result

Single threshold test

c 266395 confidence interval (2656 2692)

F value 33998P value 0001

Double threshold test

Γ1 244595 confidence interval (1345 3223)



Triple threshold test



Basic information

Number of iterations 450Search points 150

Time dimension 15Number of sections 30Number of samples 450

Table 3 reshold estimation regression results

Variable Coefficient P valueres (finlt 2431) 0027lowastlowastlowast 0002res (2431lt finle 2445 minus0018lowastlowastlowast 0001res (2445lt finle 2625) 0033lowastlowastlowast 0000res (fingt 2625) minus0019 0334inv 0088 0238edu 0012lowastlowastlowast 0004tech 0134lowastlowastlowast 0001cor minus0223lowastlowastlowast 0000open 0033 0349Note lowastlowastlowastSignificant under 1 condition


of high financial development the relationship betweenresource abundance and the level of economic developmentis positive and this stage has changed from ldquoresource curserdquoto ldquoresource blessingrdquo When finance is in the developedstage (fingt 2 625) the regression coefficient of res is aninsignificant negative number indicating that there is aninsignificant inverse correlation between economic growthrate and resource abundance Although the relationshipbetween economic development and resource abundance isnot significant at the stage of financial development theregression coefficient is negative indicating that there maybe a ldquoresource curserdquo phenomenon

To sum up first when the level of financial developmentis not high economic growth is mainly affected by theabundance of resources e economic development level ofresource-based regions at this stage is higher than theeconomic development level of non-resource-based regionsand thus the ldquoresource blessingrdquo effect occurs [36 37]Second when finance is overdeveloped the restriction ofresource endowment on economic growth decreases andthe impact of resource endowment on economic growth caneven be ignored We think there are three possible reasons(1) e transformation effect of economic developmentEconomic development is accompanied by the transfor-mation and upgrading of the industrial structure ustraditional industrial enterprises that depend on naturalresources for survival gradually withdraw from themarket astechnology changes nonrenewable resource developmentand use costs increase and market demand changes Inaddition the contribution of traditional resource-basedindustrial enterprises towards economic growth declines (2)e ldquoMatthew effectrdquo regarding capital profit-seeking can bedescribed as follows [38 39] with the development of fi-nance financial capital is sufficient and funds are invested inemerging industries At the same time when the level offinancial development increases a virtual economy is alsogenerated and its high return rate inevitably prompts fi-nancial institutions to invest a lot of funds in the virtualeconomymarket which not only promotes the generation offinancial bubbles but also leads to difficulties in financingthe real economy thereby inhibiting economic growth (3)e substitution effect of financial development During theperiod of excessive financial prosperity the role of financialdevelopment in promoting economic growth makes up forthe restraining effect of resource endowments on economicgrowth In other words in the highly developed financialstage even if there is a ldquoresource curserdquo phenomenon it doesnot show e regression results also show that the level offinancial development can promote the healthy developmentof the economy only in a reasonable range At this time theinput of natural resources including the input of physicalcapital will reach the highest utilization rate which cansignificantly promote economic growth

5 Conclusions and Recommendations

51 Conclusions Aiming at the ldquoresource curserdquo hypothesisthis article first derived the ldquoresource curserdquo and ldquofinancialthreshold effectrdquo that may exist in China en using data

from 30 provinces from 2004 to 2018 as a research samplelinear regression and the non-dynamic panel thresholdmodel were used to analyze the financial threshold effect interms of the ldquoresource curserdquo hypothesis At the same timethe paper divides the level of financial development to verifythe robustness of the conclusions of this paper e con-clusions are as follows

(1) e research in this paper shows that there is acertain correlation between the degree of resourceabundance and economic growth Whether it is aldquocurserdquo of resources or a ldquoblessingrdquo of resources issignificantly related to the degree of financialdevelopment

(2) In areas where the level of financial development islow the regions with higher resource abundance aremore dependent on resources for economic devel-opment and they can promote economic growththrough resource development

When the level of financial development is furtherimproved the degree of resource abundance cannot sig-nificantly promote economic development resulting in aldquoresource curserdquo phenomenon As the level of financialdevelopment increases the further dependence on resourcesreduces the efficiency of resource utilization and hinderseconomic growth thus there is a ldquocurserdquo of resources

When the level of financial development is high there isa positive correlation between the level of resource abun-dance and the level of economic development during theperiod of high levels of financial development e effects ofeconomic transformation at this stage are greater than thenegative effects of resource developmentus the ldquoresourcecurserdquo turns into a ldquoresource blessingrdquo

When finance is in a developed stage there is an in-significant reverse correlation between the economic growthrate and resource abundance Although the relationshipbetween economic development and resource abundance inthe financially advanced stage is not significant the re-gression coefficient is negative indicating that there may be aldquoresource curserdquo phenomenon

52 Recommendations It can be seen that the level of fi-nancial development can promote the healthy developmentof the economy only within a reasonable range At this timethe input of natural resources including the input ofphysical capital will reach the highest utilization rate whichcan significantly promote economic growth At presentmost of Chinarsquos traditional resource-based provinces arelocated in less developed regions and the level of financialdevelopment in these regions is generally lowerefore theldquoresource curserdquo phenomenon is common e research inthis article demonstrates that in order to solve the ldquoresourcecurserdquo problem and improve the level of resource-basedregional economic development we need to make financialdecisions such as promoting the development of the fi-nancial industry improving the ability and level of financialservices related to the real economy and controlling fi-nancial capital flow e details are as follows


(1) Reasonably adjust the central and local financialresource allocation and responsibilities and improvethe public financial system e increase in localgovernment financial power and the increase ingeneral power are the reasons for the increase inextra-budgetary land fiscal revenue and the rapiddevelopment of the virtual economy erefore wemust improve the public financial system rationallyadjust the central and local financial resources andresponsibilities and improve the central and localfinancial power us the mechanism of equili-brating the powers associated with different affairsfundamentally eliminates the inherent incentives forlocal governments to be enthusiastic about land fi-nance and to promote the rapid development of thevirtual economy As regards policy regulation on thereal estate market [40] fiscal decentralization pro-motes the development of the real estate industrycausing a large amount of financial capital to flowinto the real estate market is causes excessiveshort-term prosperity which leads to excessivespeculative demand for investment while the supplylags causing short-term supply shortages and risinghousing prices Deviating from its own value createsa real estate bubble so it is necessary to limit theoverdevelopment of the real estate industry (in ad-dition to policy housing such as affordable housing)reduce its virtual component and prevent the bubblefrom bursting into crisis [41]

(2) Unblock the channels for finance to enter the realeconomy First develop city commercial banksvillage and town banks financial investment holdinggroup companies and financial leasing companiesand continuously improve the financial institutionsystem [42] Second we must develop a multi-levelcapital market system continuously improve thestock market securities market expand the insur-ance market service area gradually establish regionalenergy trading spots and future markets activelydevelop derivatives trading markets promote thecontinued expansion of regional equity markets andexpand the financing channels of physical enter-prises [43] ird we must guide the healthy de-velopment of private finance is can be done byestablishing loan registration service centers in areaswhere private capital is active to promote theldquosunshinerdquo of private financing gradually incorpo-rating private financing subjects into the creditreporting system to reduce credit risk establishingand improving a private financing monitoring sys-tem to resolutely curb deception and illegal fund-raising to guide private capital to serve small- andmedium-sized private enterprises en financialinstitutions should change the single evaluationcriteria of performance as the king include servingsmall- and medium-sized microenterprise entitiesserve the development of the local real economy intothe evaluation system and reasonably guide the flow

of financial capital [44] Finally we should step upour efforts to crack down on illegal behavior in thefinancial market establish and improve local fi-nancial supervision systems and reasonably protectthe real economy

(3) e reason why the massive inflow of resource-basedregional capital into the virtual economy is impor-tant is that the operating environment of the realeconomy is not good the cost of money is high theprofits are thin and the investment channels arenarrow On this basis resource industries in resource-based areas should be de-capacified and enterprisesshould accelerate transformation promote efficientand clean development and utilization of resourcesand encourage the development of high-tech industriesand strategic emerging industries

6 Limitation and Outlook

On the basis of the previous research results this paper usesthe panel threshold model to analyze the impact of financialdevelopment on the ldquoresource curserdquo and ldquoresource bless-ingrdquo phenomena As a result of the limitations of the authorrsquosacademic level and ability and the difficulty of data col-lection this study has various limitations For example fi-nancial development is a long-term process and whether itproduces a ldquoblessingrdquo or ldquocurserdquo should be studied over alonger period of time to investigate the entire life cycle ofnatural resource development and utilization Otherwiseonly a local conclusion can be drawn For reasons pertainingto the data this article could not investigate the impact offinancial development over a longer period is will be thedirection of future research

It is difficult to obtain ideal sample data which inevitablyleads to limitations related to the research conclusion Webelieve that with the further deepening of the related re-search and the further enrichment of economic statisticsacademia will surely draw more detailed and reliable re-search results to make up and correct the deficiencies anderrors in this article Finally the ldquoresource blessingrdquo andldquoresource curserdquo phenomena may be affected by the financialdevelopment of neighboring regions as well as other spatialeffects e authors will consider the effects of spatial factorsin future research in the hope of enriching the existingresearch results

Data Availability

All data models and codes generated or used during thestudy are included within the submitted article

Conflicts of Interest

e authors declare no conflicts of interest

References

[1] S Tiba and M Frikha ldquoe controversy of the resource curseand the environment in the SDGs background the Africancontextrdquo Resources Policy vol 62 no 8 pp 437ndash452 2019


[2] E Papyrakis and R Gerlagh ldquoe resource curse hypothesisand its transmission channelsrdquo Journal of ComparativeEconomics vol 32 no 1 pp 181ndash193 2004

[3] R M Auty ldquoIndustrial policy reform in six large newly in-dustrializing countries the resource curse thesisrdquo WorldDevelopment vol 22 no 1 pp 11ndash26 1994

[4] T T Deng Y R Liu and M L Ma ldquoTourism industrydependence and total factor productivity growth a test basedon the ldquoresource curserdquo hypothesisrdquo Tourism Science vol 33no 1 pp 1ndash17 2019

[5] M J Vijge R Metcalfe L Wallbott and C OberlackldquoTransforming institutional quality in resource curse contextsthe extractive industries transparency initiative in MyanmarrdquoResources Policy vol 61 no 6 pp 200ndash209 2019

[6] P C Wang and B X Sun ldquoInfluence of energy endowmentand trade opening on resource green utilization efficiencyrdquoJournal of Dalian University of Technology Social ScienceEdition vol 2 pp 9ndash16 2019

[7] S M Pendergast J A Clarke and G C Van KootenldquoCorruption development and the curse of natural re-sourcesrdquo Canadian Journal of Political Science vol 44 no 2pp 411ndash437 2011

[8] J D Sachs and A M Warner ldquoNatural resource abundanceand economic growthrdquo NBER Working Papers vol 81 no 4pp 496ndash502 1995

[9] J-P C Stijns ldquoNatural resource abundance and economicgrowth revisitedrdquo Resources Policy vol 30 no 2 pp 107ndash1302005

[10] J G Weber ldquoA decade of natural gas development themakings of a resource curserdquo Resource and Energy Economicsvol 37 no 3 pp 168ndash183 2014

[11] T Gylfason ldquoNatural resources and economic growth what isthe connectionrdquo Social Science Electronic Publishing vol 53no 8 pp 48ndash66 2001

[12] B Q Wang and J Y Li ldquoAn empirical study on the ldquoresourcecurserdquo effect of Chinarsquos coal citiesrdquo Statistics and Decisionvol 10 pp 121ndash125 2019

[13] S H Lu ldquoParadox of the development of resource countriesfrom the perspective of institutional transmission ldquoresourcecurserdquo and ldquopath dependencerdquordquo Academic Exploration vol 8pp 59ndash66 2019

[14] P F Peretto ldquoResource abundance growth and welfare aschumpeterian perspectiverdquo Journal of Development Eco-nomics vol 97 no 1 pp 142ndash155 2012

[15] K Bjorvatn M R Farzanegan and F Schneider ldquoResourcecurse and power balance evidence from oil-rich countriesrdquoWorld Development vol 40 no 7 pp 1308ndash1316 2012

[16] J A Robinson R Torvik and T Verdier ldquoPolitical foun-dations of the resource curse a simplification and a com-mentrdquo Journal of Development Economics vol 79 no 2pp 194ndash198 2006

[17] J Song and W H Li ldquoCountermeasures for the industrialtransformation of resource-based cities in heilongjiangprovince based on the discussion of ldquoresource curserdquo areasrdquoCommercial Economy vol 9 pp 7-8 2019

[18] M Alexeev and R Conrad ldquoe elusive curse of oilrdquoReview of Economics and Statistics vol 91 no 3pp 586ndash598 2009

[19] S Haber and V Menaldo ldquoDo natural resources fuel au-thoritarianism a reappraisal of the resource curserdquo AmericanPolitical Science Review vol 105 no 1 pp 1ndash26 2011

[20] K N Xu and J Han ldquoe resource curse effect of Chinarsquosregional economy another interpretation of regional dis-parityrdquo $e Economist vol 6 pp 96ndash102 2005

[21] H-P Sun W-F Sun Y Geng and Y-S Kong ldquoNaturalresource dependence public education investment and hu-man capital accumulationrdquo Petroleum Science vol 15 no 3pp 657ndash665 2018

[22] A Y Jiang and B Liu ldquoResearch on resource developmentand economic growth in central Asia-an empirical analysisbased on the ldquoresource curserdquo hypothesisrdquo Exploration ofEconomic Problems vol 5 pp 34ndash43 2019

[23] H-P Sun W-F Sun Y Geng X Yang and B K EdziahldquoHow does natural resource dependence affect public edu-cation spendingrdquo Environmental Science and Pollution Re-search vol 26 no 4 pp 3666ndash3674 2019

[24] Y Ma and D J Cheng ldquoe gospel of resources or the curse ofresources an empirical study based on the threshold panelmodelrdquo Finance and Trade Research vol 1 pp 13ndash25 2017

[25] J X Wan and S Y Wang ldquoCan social capital and techno-logical innovation break the resource curse a study based onthe panel threshold effectrdquo Economic Research vol 12pp 66ndash89 2016

[26] L Yan ldquoServing the west of the silk road economic beltunderstanding the curse of resources and understanding thespace valuerdquo Journal of Lanzhou University vol 2 pp 127ndash138 2017

[27] R A Badeeb and H H Lean ldquoFinancial development oildependence and economic growthrdquo Studies in Economics andFinance vol 34 no 2 pp 281ndash298 2017

[28] H Yue and H J Zhang ldquoFinancial development resourcecurse and economic growthrdquo Journal of East China NormalUniversity vol 6 pp 138ndash150 2019

[29] D Y Peng C Liu and Y F Zhou ldquoStudy on the thresholdeffect of the impact of urbanization development on hazepollution in the yangtze river economic belt based on theperspective of residentsrsquo consumption levelrdquo Finance andEconomy vol 8 pp 36ndash42 2015

[30] B E Hansen ldquoreshold effects in non-dynamic panelsestimation testing and inferencerdquo Journal of Econometricsvol 93 no 2 pp 345ndash368 1999

[31] B E Hansen ldquoInference when a nuisance parameter is notidentified under the null hypothesisrdquo Econometrica vol 64no 2 pp 413ndash430 1996

[32] B E Hansen ldquoSample splitting and threshold estimationrdquoEconometrica vol 68 no 3 pp 575ndash603 2000

[33] C S Zhang and B Z Zhang ldquoe mystery of the decline inChinarsquos industrial investment rate from the perspective ofeconomic financerdquo Economic Research vol 12 pp 34ndash482016

[34] G Nahom ldquoFDI technology spillovers in the mining industrylessons from South Africarsquos mining sectorrdquo Resources Policyvol 62 pp 463ndash471 2019

[35] K Lahiri and P Schmidt ldquoOn the estimation of triangularstructural systemsrdquo Econometrica vol 46 no 5 pp 1217ndash1221 1978

[36] X J Yao and J Dang ldquoResearch on the influence of naturalresources on economic growth from the perspective of in-dustrial structurerdquo Journal of Xirsquoan Shiyou University vol 6pp 21ndash29 2019

[37] Y P Chen J He and C L Zhong ldquoldquoGospelrdquo or ldquocurserdquoresearch on the asymmetric impact of resource abundance onchinarsquos regional economic growthrdquoMacroeconomic Researchvol 11 pp 139ndash152 2018

[38] T E Ba ldquoAnalysis of the ldquomatthew effectrdquo of financial ag-glomeration on the distribution of regional financial re-sourcesrdquo Journal of Inner Mongolia University forNationalities vol 43 no 1 pp 85ndash89 2017


[39] W Y Han ldquoAn empirical analysis of the efficiency of financialresource agglomeration in economic growth in provincialregionsrdquo Business Times vol 5 pp 160ndash162 2016

[40] X L He and Y Ye ldquoResearch on the relationship betweenenvironmental pollution and regional economic growth basedon the perspective of resource curserdquo Journal of HubeiUniversity of Economics vol 4 pp 20ndash31 2019

[41] J Chen Y P Liu and L J Wang ldquoResearch on couplingcoordination development for photovoltaic agriculture sys-tem in Chinardquo Sustainability vol 11 no 4 p 1065 2019

[42] L Feng ldquoAnalysis on the transformation and innovation oforganizational capacity building from the perspective of re-source curserdquo Leadership Science vol 12 pp 31ndash34 2019

[43] Z X Tan and G N Sun ldquoe foundation of high-qualitytourism development an analysis of the gospel of tourismresourcesrdquo Social Scientist vol 5 pp 108ndash115 2019

[44] L J Wang Y Wang and J Chen ldquoAssessment of the eco-logical niche of photovoltaic agriculture in Chinardquo Sustain-ability vol 11 no 8 p 2268 2019


Abundant natural resources should be the source ofmomentum for economic growth but the ldquotrapsrdquo in eco-nomic development brought about by them have arousedmuch interest especially regarding their theoretical mech-anisms [4] With the increasing amount of research on thetransmission mechanism there are fewer studies focused onsolving the ldquoresource curserdquo Why do resources cause de-velopment traps Exploring the cause of the ldquoresource curserdquoand overcoming resource problems have become an im-portant issue for many resource-abundant countries [5]

China is a big resource-rich country Many resource-rich provinces and cities have contributed to the countryrsquosconstruction However with the exploitation of resourcesmany resource-based areas have encountered developmentalobstacles e questions of how to solve resource-basedproblems and promote resource-based areas are of greatsignificance in terms of helping its transformation [6]Evidence shows that resource abundance will also have anegative impact on the development of finance but moststudies do not mention the relationship between the ldquofi-nancial development levelrdquo and ldquoresource abundancerdquoWhen sorting through the development context of resource-rich countries we found that not all resource-rich countrieshave fallen into a development trap Compared with thetragic fate of some countries that have been ldquocursed byresourcesrdquo other countries have achieved considerable de-velopment by virtue of their resource advantages Researchby Pendergas et al [7] found that ldquoresource countries thathave gotten rid of the development trap have developed theirfinancial development on the contrary resource countriesthat have suffered from curses often have very low levels offinancial developmentrdquo Is this a coincidence Finance is thecore of the modern economy and an effective financialsystem has the important function of rationally allocatingresources and vigorously promoting economic develop-ment Can it also be applied to natural resources On thebasis of the financial perspective this paper explores how thelevel of financial development affects the resources curse Onthe basis of the panel data from 15 resource-based regions inChina from 2004 to 2018 the Hansen threshold panel modeland the quasi-irrelevant regression estimation (SUR)method were used to empirically analyze whether the fi-nancial development of resource-based regions would affectthe resource curse

e rest of the paper is organized as follows Section 2 is aliterature review which mainly introduces the research ofdomestic and foreign scholars in this field and provides atheoretical reference for the research of this article Section 3presents a detailed description of the method used in ourstudy Section 4 presents and discusses the results from theempirical analysis Section 5 summarizes the conclusionsand puts forward corresponding countermeasures andsuggestions based on the empirical analysis results

2 Literature Review

e introduction of the ldquoresource curserdquo hypothesis hassparked widespread discussion in academia Foreign re-searchers have previously studied the ldquoresource curserdquo

problem Most of the research conclusions support theldquoresource curserdquo hypothesis with the ldquoresource curserdquotransmission mechanism being proposed from differentdimensions studies [8ndash10] Gylfason [11] systematicallyexpounded on how the transmission mechanism of theldquoresource curserdquo affects economic growth and found thatthere are four transmission paths including ldquoDutch diseaserdquorent-seeking government decision-making errors and hu-man capital investment Subsequently a large number ofscholars enlarged their assumptions on the analysis of theldquoresource curserdquo transmission mechanism For examplePapyrakis and Gerlagh [2] found the existence of threeldquoresource curserdquo transmission mechanisms governmentcorruption foreign trade and education investment Wangand Li [12] used recent panel data at the provincial andmunicipal levels and coal development intensity as ameasurement index to test the ldquoresource curserdquo effect ofChinarsquos coal citiese results show that the coal resources ofthe sample coal cities were significantly positively correlatedwith economic growth which leads to the conclusion thatthe ldquoresource curserdquo is not significant at the city level inChina Lu [13] believes that the root cause of a resource-richcountryrsquos long-term low performance in terms of devel-opmental path is not based on the ldquocurserdquo of abundantnatural resources but should be attributed to the institu-tional conditions and the ldquopath dependencerdquo caused bychanges At the formal regulatory level it is difficult forcountries to benefit from an excess of knowledge and skills ina participation structure associated with resource propertyrights which is further constrained by the dilemma ofproduction capacity and operation and management Interms of constraints on the informal level under thestructure of a rent-seeking culture fluctuations in the re-source cycle weaken the systemrsquos reason distort the in-centive structure and induce institutional changes At thelevel of implementation path the unpredictability of weakintertemporal commitments raises the transaction costs ofthe government entering into and fulfilling contracts withdevelopmental enterprises Peretto [14] expounded on thetransmission mechanism of the ldquoresource curserdquo from theperspective of technological innovation and found that therewas a significant negative correlation between the abun-dance of resources and regional technological innovationcapabilities Bjorvatn et al [15] found that rent resources andpolitical resources are a significant factor for the ldquoresourcecurserdquo and thereafter the transmission mechanism of po-litical resources and political systems became the focus ofresearch by experts and scholars [16] However as towhether it is the ldquoresource blessingrdquo or ldquoresource curserdquo theacademic community has always been undecided and hasnever come to reach a consensus Studies by Song and Li [17]and Alexeev and Conrad [18] all challenge the ldquoresourcecurserdquo hypothesis ey analyzed and found that the re-source endowments of Russia Canada Norway and othercountries have a significant positive effect on economicgrowth At the same time they also pointed out that there isno direct evidence that rich natural resources have a de-structive effect on economic growth Haber and Menaldo[19] by excluding regional heterogeneity and time





3 Method














(5)



Sn(c) (6)




Sn(1113954c) (7)





Sn(1113954c) (8)



4 Results Analysis





















(10)




(11)











(12)


















Basic information

























Data Availability




References



















































3 Method














(5)



Sn(c) (6)




Sn(1113954c) (7)





Sn(1113954c) (8)



4 Results Analysis





















(10)




(11)











(12)


















Basic information

























Data Availability




References




















































(5)



Sn(c) (6)




Sn(1113954c) (7)





Sn(1113954c) (8)



4 Results Analysis





















(10)




(11)











(12)


















Basic information

























Data Availability




References





























































(10)




(11)











(12)


















Basic information

























Data Availability




References






















































(12)


















Basic information

























Data Availability




References



































































Data Availability




References
















































doesfinancialdevelopmenthamperorimprovetheresource curse

Documents