By using a panel data set covering 10provinces in China from 1986to 1999,we present the stylized facts of rural taxation in China.We find that the rural taxation rates did not increase as fast as many people conceived.The main reason that rural taxation became an acute problem in 1990s is in large part due to the increase of rural income disparity after 1990s and the uneven tax and fee distribution among different income groups.The low-income peasants pay much higher shares of taxes and fees as percentages of their incomes,which was mainly due to the fact that the poor people are usually the group of people with lowest proportion of income from non-agricultural sources,thus they are more vulnerable to rural taxation.We also find that government regulation is an important factor to explain rural taxation burdens.
The reason is that the regulation entails a lot of costs not only in policy implementation and administration,but also in corruption,since local officials can impose extra charges on peasants in the name of implementing the central government regulations,given that the local enforcement cost of regulation is unobservable to central government.Most importantly,with a general theoretical framework,we further analyze the relationships between government regulation and rural taxation,rural factor mobility,and also rural income growth and disparity.We hypothesize that with differentiating regulation such as grain procurement across regions (and also across households),regions (and households)that are more heavily regulated tend to have heavier rural taxation,more vulnerable to local bureaucracy expansion and serious corruption,lower labor and land mobility,and thus,will result in lower income growth.Preliminary empirical studies strongly support our hypothesis.Moreover,the more heavily regulated regions (and households)will be more locked in agricultural production,which will further result in heavier rural taxation and even lower income growth.This constitutes a vicious cycle for the more heavily regulated regions (and households),while the opposite happens to the less regulated regions (and households).
In conclusion,our general framework can help us to understand the increasing income disparity,the highly regressive nature of rural taxation and their relationships with government regulation simultaneously.Policy implications are also drawn in rearranging inter-governmental relationships,removing or relaxing of government regulations,and more generally,establishing a more reasonable rural public finance system.
In recent years,the problem of rural taxation in China,especially of increasing local informal charges on peasants,has become more acute.The central government has been aware of the problem for a decade,and has been taking various steps to alleviate the problem.Up to now,different state ministries have promulgated 37official regulation documents to manage and reduce peasant tax burdens.Yet to date,these actions have met with limited success.
Many scholars have paid attention to the problem of rural taxation in China.Most researchers,especially some political scientists focused on the nature of political system as the main explanation for the surging fees on peasants,and proposed to reform the current rural political governance by promoting rural election and advocating rural local governance (Bernstein &Lu,2001;Qing 2000;Cao 2001).Others argued that the lack of financial resources and low share of local government budget in total fiscal budget revenue led to higher rural tax burdens (Yi,2000;Cao,2001).Notwithstanding the wide attention given to rural tax burden,there has not been any systematic research with empirical support on this issue,and it is far away from clear to us for exploring the reasonable explanation.
This paper is the first step in a systematic study on the problem of rural taxation,and more generally,the problem of the relationships between government regulations and rural taxation,rural factor mobility,and also rural income growth and disparity.Part I gives a brief introduction to the rural taxation system in China.In Part II,we will present the stylized facts of rural taxation by using a large panel dataset at rural household and village level for 10provinces from 1986to 1999.We find that the average level of rural taxation relative to rural net income after 1990s did increase in most provinces,but not as fast as most people conceived.We also find after 1990s rural taxation became more and more regressive mainly due to increasing income disparity in rural areas and the uneven tax and fee distribution among different income groups.In Part III,we present a general theoretical framework to approach the problem of rural taxation.We argue that the un-funded government regulations such as grain procurement and birth control constitute the key to for us to understand the problem of rural taxation and more generally,the problem of rural income growth and disparity.By focusing on the regulation of government grain procurement,we hypothesize the heterogeneous enforcement regulation across regions (and also across households),regions (and households)that are more heavily regulated tend to have heavier rural taxation,more vulnerable to local bureaucracy expansion and serious corruption,lower labor and land mobility,and thus,will result in lower income growth.In Part IV,we present preliminary empirical studies that support our
hypothesis.In part V,we conclude that differentiating enforcement of the same government regulation will lead to higher income disparity and differentiating rural taxation burdens in rural areas,which helps us to understand the increasing income disparity,the highly regressive nature of rural taxation and their relationships with government regulation simultaneously.Possible policy implications are also addressed in Part V.
Part I:The Current Rural Taxation System in China
The overwhelming majority of China's nearly 1.3billion people live in the rural areas of the country.Prior to the 1980's,agriculture and rural enterprises were collectively owned and directed by the state or a local collective.Government policies favored urban residents,who enjoyed guaranteed employment,free housing through their work units and free health care,while rural areas were heavily regulated not only in factor mobility,but also in basic production and distribution decisions.Through a system of household registration and residence permits,citizens in the rural areas were severely limited in their ability to move to urban areas.Decisions such as what to produce,agricultural pricing and even income distribution within villages are all mandated from governments above according to a rigid set of rules and regulations.Agricultural sales were conducted between rural collectives and the state.At that time,rural areas were systematically deprived of resources above subsistence level by the government through implicit taxation such as the “price scissor”between industrial and agricultural products.In another word,rural taxation was collected directly from collectives through agricultural input and product pricing directed by government,and thus was implicit in nature.
Since late 1970s,although still heavily favoring the cities,the Chinese government adopted a number of reforms to encourage economic growth in the rural areas.One aspect was to de-collectivize agriculture and give the peasants greater autonomy in managing their production.This became known as the responsibility system.Instead of working the collective plots,peasants were given the right to lease specific land for 15-year terms.These lease terms have subsequently been expanded to 30years and may soon increase to 50years.One aspect of the responsibility system was to shift the basic unit of rural taxation from the collective to the individual household.With this shift,households became individually liable for various taxes and most fees.For example,by having the right to cultivate a given plot of land,the household became responsible for taxes and fees on the land itself as well as the production from the land.At the same time,the government gradually liberalized the prices for agricultural products,but still kept some regulations on grain production and procurements requiring peasants sell part of their grain to government.With the shift of taxation units from collectives to households,gradual liberalization of agricultural pricing and regulations on some major agricultural products,rural taxation became more explicit.
Rural taxation in China is unique in the sense that peasants not only need to pay formal taxes,but also need to pay informal fee charges.In rural areas,formal state taxes are levied on agricultural and non-agricultural operations;however,since only part of the state taxes are retained for local governments
and these retained revenues only cover part of the local government expenditure,local governments at the county and township level,and community organizations at the village level have to fill the financial gap by levying various fees on rural households.
1.1Rural Taxation:Basic Classification
The typical direct taxes and fees imposed on peasants are listed below.
Formal rural taxation includes agricultural tax,special product tax,slaughter tax,farmland utilization tax and various business taxes.
(1).Agricultural tax is levied on peasants based on grain output.It is a proportional tax.The national average tax rate is about 2.5%at present.According to an estimate by the Ministry of Finance,in the Ninth Five-Year Plan period,the average annual agricultural tax in the whole nation is about RMB 25.4billion.(State Planning Commission,Internal Report,2000).(2)Another major tax over agriculture is the special agricultural product tax.Since mid-1980s,with the development of cash crop and aquaculture,the agricultural structure has seen significant changes with the proportion of cash crop plantation,forestry,fishing and animal husbandry increasing rapidly.To distribute tax burdens on different sectors within agriculture evenly,the “Provisions On Agricultural Special Products Tax”was promulgated in 1994.The tax is levied over tea,fruits,silkworm,animal husbandry and aquaculture products.In 1999,the special products tax reached RMB 8.89Billion,an obvious increase from RMB 7.96billion in 1996,an annual increase of 3.6%.
(3)The government also levies slaughter tax,farmland utilization tax and various rural business taxes.The slaughter tax is levied on the number of cattle.The farmland utilization tax is levied according to the area of arable land occupied by peasants for housing and other non-agricultural construction.However,rural business taxes(mainly excise and value-added taxes)on rural non-agricultural economic activities(such as TVEs,rural transportation et al)constitute the major part in this category.According to the Ministry of Agriculture,these taxes increased rapidly,from RMB 120.6billion in 1996to RMB 144.98billion in 1999,with the annual growth rate of 6.3%.According to Ministry of Agriculture,rural business taxes should also be considered as tax burdens on peasants.It will be argued later in the text that it is misleading to include rural business taxes in the calculation of rural tax burdens.(see section 2.2).At the present,to simplify presentation,we will call all formal rural taxation except rural business taxes as agricultural-related taxes,thus it includes agricultural tax,special product tax,slaughter tax and land occupation tax.
Besides formal government taxation,the Chinese peasants also need to pay various fees to local government and village community organizations.In China,The typical revenue sources of local governments are listed below.
(1)Township-pooling funds:the function of township government is to provide basic public goods such as education,public security,law and order,civil service and carry out the state mandated tasks of family planning,and government mandatory grain procurement.Since the government at higher levels cannot provide sufficient financial resources,only parts of the township budget can be covered by rural state taxes collected.The township government needs to finance most of its budget within the governed area.The current national policies permit the township governments to levy pooling funds for the following five purposes:school,family planning,support for veterans,military training,and road construction and maintenance.Therefore,they are called “the five township pooling funds”.
(2)Village Levies:in addition to fees to township governments,peasants must also pay to the village community organization.The community organization at village level is nominally not a level of government.But in reality it is still controlled by governments at higher levels and must carry out the tasks mandated by the government.The current policy promulgates that village community organization can charge collective accumulation fees,collective welfare fees and administration fees,the so-called “Three Village Fees”.The collective accumulation fees are used for rural water conservancy construction,purchase of productive fixed assets and establishment of collective enterprises;collective welfare fees are used to support families with special difficulties and other collective welfare.Administration fees are used for cadre compensation and administration costs for village organization.In addition,rural households must pay the so-called contracting fee,the collective fee for the use of assigned land.
(3)Fees,assessments and local fundraising:these are mainly fees charged on peasants without explicit government regulations or legislation.There are sometimes tens or even more than one hundred items,ranging from charges for road and school construction and other local improvement projects,purchase of insurance,to charges for marriage certificates,housing construction,to prohibitive prices for electricity and tap water,etc.In some regions,local government and village community organizations even force peasants to borrow money to pay the various fee charges,and also transfer public or collective debt to individual households.In some regions,marriage certificates cost RMB 600Yuan,a quarter of the peasant’s average annual net income (State Planning Commission,Internal Report,2000).
(4)Fines:fines are collected by various government agencies and localities for various infractions,such as birth control violations.The government sets very strict regulation on birth control and imposes very high fines on infractions.However,in many cases,peasants can still have more children if they pay fines or send bribes to local officials.In fact,one way for local governments to cover revenue shortfalls is the arbitrary imposition of large fines and penalties for minor violations of rules (State Planning Commission,Internal Report,2000).
In recent years,the township pooling funds and village levies have increased significantly.The central government has mandated that these levies cannot be more than 5%of peasants’net income.However,the situation is not under control in some regions.Some local governments deliberately exaggerate peasants’net income and levy the funds and fees according to the falsely reported income.According to a report (State Planning Commission,2000),from 1993to 1998,the national township pooling funds and village levies increased from RMB 39billion to RMB 73billion,with an annual growth rate of 13.9%.The amount per capita increased from RMB 44.6Yuan to RMB 84Yuan,with annual growth rate of 13.8%.See Table 1below.
Table 1All Township and Village Informal Levies:1993--1998
Other than the explicit fees listed above,peasants also need to take the following implicit burdens:
(5)Monetary equivalent:this is the corvee labor services,primarily provided
by male labors--e.g.labors for flood prevention,maintaining and expanding irrigation systems,road and school construction,water conservation and afforestation projects.In recent years,these corvee labor services have increased rapidly.In 1994,the average labor in rural regions contributed 16.4working days and in 1999,the number increased to 18.2working days,in some years it even reached 23working days.The cash equivalent per capita in 1998reached 89.3Yuan,130%higher than 1994level.In addition to that,some local governments take advantage of the national corvee labor policy,requiring peasants to contribute corvee labor services,with the real intention to ask rural households to substitute cash for labor-the so called “Cash for labor”(State Planning Commission,Internal Report,2000).In 1999,the total amount reached 6.4billion,with 6.9Yuan per capita and 13.46Yuan per labor (this is not included in Table 4).
(6)Educational charges:different from urban residents,rural residents must pay for most of their education.Although the national policy mandates that the 9-year primary and secondary education is compulsory,government cannot provide sufficient resources for rural residents to fulfill this mandate.To fill the financial gap,many schools deliberately violate government policies and charge different fees other than tuition according to arbitrary standards,such as electricity and water fees,experiment fee,library fee,school policing fee,examination fee,etc.In Hubei province,it was found that there are 59charges beyond government policy.
(7)Hidden burdens.This includes compulsory sale of grain to the government at below market prices.Depending on the region,the grain may be in the form of rice,maize or wheat.There are also various user fees charged at above market level such as water,electricity fees.
The uniqueness of the rural taxation system lies in the fact that a large part of the village and township government expenditure comes from direct taxation,especially the informal rural fee charges on peasants that do not enter into formal government budget.On the one hand,township level governments and community level organizations are responsible for local public goods provision;on the other hand,they must fulfill a lot of tasks mandated from governments at higher levels,notably the government grain procurement,birth control and compulsory education.Since these tasks are usually un-funded mandates,fee charges must be imposed on peasants to finance the implementations.Therefore,in many rural areas,the bulk of local cadres’work is to charge fees,and then use the fee charges to implement the grain procurement,birth control,compulsory education and many other tasks required by higher-level
governments.Due to the difficulties in implementing these policies,the administrative costs are usually so high that rural governments usually find they have insufficient resources and efforts to provide basic public goods.In addition,in the name of implementing the government policies,local cadres usually over-charge peasants for the purpose of self-consumption and corruption under the political system without sufficient supervision.
Table 2below shows the quantity of agricultural-related taxes and rural fees from 1994to 1998and their ratios to total formal government taxes.As we can see from the table,agricultural-related taxes is only a relatively small part of total government taxes,but the quantity of rural fees is much higher,close to twice as much as that of agricultural taxes.
Table 2Rural Taxation In Total Government Tax Revenue
Note:1These include agricultural,special product,slaughter and land Occupation Tax,but does not include various business taxes by rural secondary and tertiary sectors.
Source:1,Internal Report On Rural Tax Burdens,State Planning Commission,20002,New China Fifty Years'Government Finance Statistics 2000
1.2Definition of Tax Burdens:Some Clarifications
According to the official statistics (State Planning Commission,2000),if corvee labor services and various illegitimate charges are excluded,the average government taxes and local fee levies per capita reached RMB 281
Yuan in 1998,twice the level in 1993,with an annual growth 14.8%.This growth rate is much higher than the growth of rural net income.See Table 3below.
Table 3:Rural Tax and Fee Charges 1993-1998
Note:1These include agricultural,special product,slaughter and land Occupation Tax,but does not include various business taxes by rural secondary and tertiary sectors.
Source:1,Internal Report On Rural Tax Burdens,State Planning Commission,20002,New China’s Fifty Year Government Finance Statistics 2000
However,some caution must be taken when interpreting Table 3.In rural taxes,the agricultural tax,special agricultural tax,land utilization tax and slaughter taxes only constitute a small part of all state taxes levied in rural areas,while rural commercial taxes (such as excise tax and value-added taxes paid by rural business,for example,the TVEs)constitute the majority of rural taxes paid.From Table 3,we can see that
the figures for agricultural-related taxes are only 36.9,39.7and 39.9billion in 1996,1997and 1998respectively.This means that they constitute about 23-24%of all state taxes levied in rural areas.
We argue that it is misleading to attribute the rural business taxes to rural households and thus consider them as rural tax burdens,since these taxes are levied on rural non-agricultural businesses.This point is very important since it relates to the assessments of the degree of rural burdens.We will calculate the “direct taxation”on rural households.In rural areas,personal income tax can basically be ignored due to the difficulties in levying the tax in rural areas,the low-income level of rural households,and also the characteristics of rural income sources (largely in-kind incomes such as agricultural output).However,the governments impose agricultural tax,special product tax,slaughter tax and land occupation tax directly on rural households.The following fact needs to be kept in mind:different from urban households which is mainly a unit of consumption,rural household is not only a unit of consumption,but also a basic unit of agricultural production.Therefore,these agricultural-related taxes can be considered as direct taxation on rural households.However,the non-agricultural taxes,such as the rural business taxes,should not be considered as rural household tax burdens,since as in urban areas,these taxes are paid by enterprises rather than households.Therefore,in our calculation of rural direct tax burdens,we will not consider the rural business taxes as tax burdens on rural households.Therefore,the numbers in the last row of Table 6indicating that rural direct tax burdens as a percentage of rural net income was 15.3%in 1993and was about 13%in 1997and 1998are apparently over-estimations of the real situation.
One important point to note is related to tuition.One major difference between urban and rural household expenditure structure in China lies in the education expenditures.China has a national 9-year compulsory education system.In rural areas,the nine-year compulsory education is mostly financed by rural households,while much of the educational expenses in urban areas is financed or subsidized by local government.In urban areas,
city governments usually pay for most of the school basic construction and teacher compensation through government budget.While urban households pay much less proportion of their income in tuition,they usually enjoy education with much better quality.In this sense,relative to urban residents,rural residents pay much more for the primary and secondary education.Therefore,the rural tuition and various educational charges can
be considered,in a large part,as a tax on rural households.I will consider this factor in my calculation of urban and rural tax burdens.
Part II Direct Taxation on Rural Households
We will study direct rural taxation using the rural household survey data covering over 6000rural households and 120villages in 10provinces in the years of 1986-1991,1993,1995-1999.Estimates of the distribution of the direct rural tax burdens among different income groups for the whole sample will also be provided.The data set is from an annual Fixed Point Survey carried out by the Ministry of Agriculture in China (see Appendix I).The dataset covers 10provinces,Shanxi,Jilin,Jiangsu,Zhejiang.Anhui,Henan,Hunan,Guangdong,Sichuan.Within the 10provinces,Zhejiang,Jiangsu and Guangdong are in the coastal areas;Henan,Hunan,Jilin,Ahhui are in central China;while Gansu and Sichuan are in west China.Among the 10provinces,Sichuan,Hunan,Jiangsu and Jilin are high grain-producing provinces.
Using the data from the rural household survey.We calculate the tax burdens for rural households in the 10provinces.The graphs describing the dynamics of rural tax burdens for the ten provinces from 1986to 1999are presented below.Several definitions of tax burdens are given:Fee1is defined as all agricultural-related taxes (including agricultural tax,special agricultural product tax,slaughter tax)plus all township and village levies as a percentage of household net income.Fee2is defined as township and village levies (including the three types of village levies,five types of township pooling funds legitimated by national government policy,plus various local charges not legitimated by national government policy as a percentage of household net income.Fee3is defined as those various charges not legitimated by national government policy but imposed by local (county or township)government and village community organizations as a percentage of household net income.Tax is the average agricultural-related taxes paid by rural households as a percentage of rural household net income.Tuition is defined as tuitions and fee charges by schools as a percentage of household net income.Totalfee is defined as Fee1plus Tuition.Therefore,Totalfee=Tuition+Fee1,Fee1=Fee2+Tax,Fee2=five types of township pooling funds+three types of village levies+Fee3.In the following graphs,we only present Totalfee,Fee1,Fee2and Fee3to simplify our presentation.According to our definition,the difference between Totalfee and Fee1is Tuition.The difference between Fee1and fee2is tax.The difference between Fee2and Fee3is legitimate township and village levies(the five types of township pooling funds and three types of village levies).We have tuition data only after 1995,thus Totalfee start from 1995.different provinces for the surveyed sample.In more developed provinces such as Guangdong and Zhejiang,tax burdens are much lower than those of less developed provinces such as Jilin,Hunan,Henan and Sichuan.By the end of 1990s,the totalfee rates are about 8-9%in Guangdong and Zhejiang,but reached 11-13%in Shannxi,Anhui,Henan,Gansu,15-17%in Sichuan,Hunan and Jilin.Jiangsu is an exception since its totalfee rate is also as high as 13%although it is a relatively rich province in China.After 1993,totalfee rates in all provinces have more or less increased.In Guangdong,it rose from 7.85%to 8.49%from 1995to 1999,while in Jilin it increased from 12.01%to 15.5%,in Sichuan from 11.41%to 15.95%,and in Hunan 12.84%to 17.9%.Much of
the increase comes from increases in tuition.In Guangdong,tuition rose from 3.93%-4.86%,while in Jilin from 4.18%to 6.35%,in Sichuan from 3.93%to 4.86%,and in Hunan from 5.95%to 9.33%.For fee1(tax plus all local fees,but excludes tuition),only Guandong had a slow decline in 1990s,in Zhejiang and Shanxi,it kept stable from 1993to 1999,in other provinces it increased at different paces.The overall judgment is that fee1did not increase very rapidly even in the provinces with higher burdens.Most of the increase of totalfee comes from increases of tuition.This is contrary to people’s general belief
that fee charges have increased very fast after 1990s,especially after mid 1990s.Then how can we reconcile the rural taxation dynamics with the fact that problem of rural taxation became more acute after 1990s.Part of the answer lies in the stylized facts presented in the next sub-section on taxation rates for different income groups.(Another important fact to note is that except in the richer provinces of Guangdong and Zhejiang,local levies (as indicated by fee2in the graphs above)usually constitute half or more than half of fee1.While in Guandong and Zhejiang,state taxes usually constitute most of the rural taxation by peasants.This means that local fees are relatively minor in rural taxation in these two richer provinces,but become a more serious problem in poorer provinces.)
2.2Direct Rural Tax Burdens For Different Income Groups.
Now we will study the taxation rates for households in different income groups for different years.I group households in all provinces into different (per capita net)income group and calculate the various taxation rates for these income groups.I draw graphs for fee1,fee2,fee3(note again,fee1=fee2+tax,fee2=township pooling funds+village levies+fee3).Since we have data on educational expense (which is larger than tuition)for the whole period from 1986to 1999,but do not have data on tuition before 1995,we cannot draw graphs for totalfee from 1986(keep in mind that totalfee=fee1+tuition).However,since we find that on average educational expenses as a percentage of rural net incomes are very close to tuition as a percentage of rural net income (the differences are only 1-2%percentage points),We define edufee as educational expenses as a percentage of rural net incomes,and tfee =fee1+edufee,so that we can also draw graphs for tfee1as appropriations to the totalfee defined in section 3.2.2and facilitate the comparison of taxation burdens for different income groups among the years beginning from 1986.Again,according to our definition,in the following graphs,the difference between tfee and fee1is edufee.The difference between fee1and fee2is tax.The difference between fee2and fee3is legitimate township and village levies (the five types of township pooling funds and the three types of village levies).
In the last section,we see that excluding tuition,fee1,fee2and fee3did increase after mid 1990s in most provinces,but the increases are not very significant.However,the problem of rural tax burdens becomes more and more apparent after 1990s,especially after mid-1990s.Why did this happen?With the graphs of tax burdens at different income groups,we can find that rural tax burdens as represented by fee1,fee2and fee3have become very regressive in 1990s,especially after 1995.For example,in 1986,fee1is 10.5%for the group with lowest income group (less or equal than RMB 200per capita),but it is about 9%for those with highest incomes (above 4000RMB per capita).In 1999,the corresponding numbers became about 25.5%and 4.4%.If we further add edufee as defined above (which is close to tuition,as defined in last section,but a little bit larger on average),situation will be even worse.In 1986and 1989,the index for total rural burdens,tfee,was only 13.7%and 23.7%for the lowest income group respectively,but reached 68.7%and 55.1%for the lowest income group in the year of 1997and 1999respectively.
Therefore,we will argue that the main reason for the dramatic change in tax incidences of different income groups is that rural income disparity became much higher in the 1990s,while the charge of rural fees did not change accordingly.A research that also used the same Fixed Point Rural Survey data but covers all 31provinces in China and more than 22,000households support this argument (Guan,Zhang and Guo,2001).According to that research,if all the households are divided into five groups of equal number of households according to income level(let us name the five groups as low-income,quasi low-income,middle-income,quasi high-income and high-income group),and if we define the average income of the middle-income group is 100,then in 1990,the ratio of incomes for the low-income,middle-income and high-income groups is 43:100:275in 1990,but reached 40:100:299in 1995.In 2000,it was furthered increased to 36:100:342.The Gini index increased from 0.39to 0.43from 1995to 2000.However,their research did not comb through the rural taxation data and give relations of income disparity with rural taxation.
We will also argue that the poor people are usually the group of people with lowest proportion of income from non-agricultural sources,thus these people are most likely subject to rural tax and fee charges and the poor households usually pay disproportional taxes.The table below does support
Table 4Income Level,Income Structure and Tax Burdens 1999
Note :1Average income is net income per capita for the specific income groups.
2Income structure is the total operating income from agriculture as a percentage of total operating income for the specific income groups.
3Fee1is total state taxes and local fees as a percentage of total net income for the specific income groups.
4Fee2is total local fees as a percentage of total net income within the specific income groups.
From the table above,we can see that the higher the income,the lower the share of peasants’income from agriculture,and the higher the tax burdens.Therefore,the argument above is valid.
Part III A General Theoretical Framework of Rural Taxation
In this section,we will carry out some economic analysis on the problem of rural taxation.As we can see from the stylized facts presented in the last section,totalfee as a percentage of rural net income did increase,but the increase mainly came from increase of tuition,or more generally,educational expenses.Conventionally,tuition is
not included in the calculation of rural tax burdens,but as we pointed out in Part I,China has a national 9-year compulsory education regulation applied both in urban and rural areas.However,basic education in rural areas is mostly financed by rural households,while much of the educational expenses in urban areas is financed or subsidized by local municipal government.Therefore,the rural tuition and various educational charges can be considered,in a large part,as a tax on rural residents.After 1990s,the nine-year compulsory education regulation has been more actively implemented in rural areas,which led to higher educational expenditures of rural households.
However,if we follow the conventional definition of rural taxation,we find that fee1(agricultural-related taxes plus rural fees),fee2(rural fees)and fee3(illegitimate rural fees)did not increase very rapidly even at the provinces with higher burdens.How can we reconcile the rural taxation burden dynamics with the fact that problem of rural taxation became more acute after 1990s?The stylized facts on taxation rates of different income groups presented in section 2.2offer a partial explanation:an important reason that rural taxation problem became more acute is due to the highly regressive nature of rural taxation.
However,this explanation is only partial for at least two reasons.First,the fact that rural income disparity became much higher in 1990s is something that needs to be explained in the first place for us to offer an integrated theoretical framework to explore the problem of rural taxation.Second,as we mentioned in Part I,in rural China,township level governments and community level organizations are not only responsible for providing local public goods,but also responsible for fulfilling a lot of un-funded mandates from central government,notably the government grain procurement,birth control and compulsory education (which can also be viewed as a type of public goods,but mandated from above).These un-funded regulations created heavy financial pressures on local government,which leads to heavy rural fee charges and at the same time insufficient rural public goods provision.Therefore,a more general theoretical framework is definitely needed to explain systematically the problem of rural taxation,and more generally,the problem of the relationship among rural taxation,government regulations and rural income growth and disparity.This part will be a first step in this endeavor.
As a matter of fact,we cannot address the problem of rural taxation in isolation from the integrated picture of rural economic development in the past two decades.The stylized facts presented in Part II show that rural taxation (at least the explicit taxation)was less serious and regressive in 1980s than in 1990s.It will be a very incomplete explanation if we simply attribute the rural taxation problem to the increasing income disparity in rural areas and the uneven taxation burdens among different income groups.A more general theoretical framework must also explain the dynamics in rural income disparity.In this section,we will argue that government regulations in rural areas constitute the key to understand the problem of rural taxation,and more generally,the problem of rural income disparity within an integrated theoretical framework (the problem of rural taxation has drawn a lot of attention
from both academia and government.Up to now,several arguments from the perspectives of inter-governmental fiscal arrangement,urban biased policy and rural election and governance have been put forward.In Appendix II,we introduce these arguments and argue that these arguments either cannot match empirical evidence,or cannot explain as much facts within a single theoretical framework).
Intuitively,the highly regressive nature of rural taxation are connected with
the following facts:first,there has been increasing income divergence among peasants after 1990s;Second,government regulations in rural areas distort rural resource allocation and lead to deviations from comparative advantages both at local and individual household level,thus tend to enlarge the income disparity among regions and households;Third,due to the information asymmetry in regulation enforcement between the central and local government,government regulations may easily lead to bureaucracy expansion and provide local cadres more opportunities in corruption,which will further aggravate rural taxation burdens and lower income growth for heavily regulated regions and households.
Besides the nine-year compulsory education,the two most important regulations in rural China are government grain procurement and birth control(see Appendix III for a more detailed explanation).The difficult tasks of government grain procurement and birth control have not been fully funded.Local governments have to balance their budget by imposing informal fee charges on peasants.In any society,local government is in a position to provide public goods such as basic law and order,transportation,and primary education.There is not much dispute that local residents should provide (at least part of)the financial resources.However,except education,these regulations from central government entail huge costs in implementation but at the same time do not contribute to rural productivity and growth as local public goods do.
The regulation argument holds that that the revenues from rural taxation,especially from the informal fee charges,and the expenditures at local level are strongly connected to government regulations and policies mandated from higher level government,especially the central government.Since much of the rural direct taxation revenue is used to implement these un-funded higher-level government regulations and policies rather than provide local public goods,thus revenues from rural taxation cannot play its roles in providing local public goods,promoting rural productivity and enhancing rural income.As we can see from the stylized facts presented in last section,rural tax burdens in different provinces differ significantly.The less developed and more agricultural provinces (especially the high grain-producing provinces)usually suffer heavier tax burdens.We argue that this is not only due to the fact that these provinces have lower incomes,but also because these provinces usually suffer more from higher-level government regulations and policies,thus local governments must tax more on peasants to implement these regulations and policies.Apparently,there is significant information asymmetry in regulation implementation between local governments and higher-level governments.Namely,given a central government regulation,the implementation and the administrative costs entailed cannot be fully
observed by higher-level government.In this case,local government may easily over-charge peasants in the name of implementing the central government regulation to increase local revenue,expand local bureaucracy by excessive staffing,and promote personal consumption by corruption of local officials.
However,there is still an important problem in our theory and also in hypothesis testing if we argue that central government regulations play important roles in the problem of rural taxation.If the central government implement the regulation homogeneously across the country,how can we explain the heterogeneous taxation rates and further,the larger income disparity across the regions in 1990s?
We argue that although the central government mandates the regulations to be implemented across the country,the enforcement,or the difficulty in enforcement of these regulations is heterogeneous across different regions.For example,the central government implemented the grain procurement policy in almost all provinces.However,the quantities of government grain procurement (and the ratio of government procurement to total grain output)are very different across different provinces (and counties,townships,villages and even households).The quantity of grain procurement (and its ratio to total grain output)for every province,county,township and village is determined by upper level governments according to a set of rules which take in to the factors such as natural conditions,historical factors,and even political concerns such as local food self-sufficiency.Apparently,these factors are not controlled by peasants or villages who sell grains to government,and thus can be viewed as relatively exogenous.Therefore,there is sufficient differentiation in grain procurement regulation enforcement across regions and even across households.For the birth control regulation,the central government policy are much more
homogeneous across regions.However,the difficulties in implementing the relatively homogeneous regulation differ significantly across different regions.In poor areas where income is low,non-agricultural employment limited and education underdeveloped,peasants usually want to have more children than their counterparts in richer regions.Therefore,the difficulties in implementing the relatively homogeneous birth control policy in poorer regions are much higher,which may entails higher administrative costs and more staffing.
With this “homogeneous regulation with heterogeneous enforcement”concept introduced,we can build up our integrated theoretical framework with the following set of logically consistent hypothesis.Due to difficulties in data and measurement,we will describe our theoretical framework on the basis of government grain procurement regulation (since in the rural survey data set,we cannot find good proxy to represent the enforcement of birth control regulation).However,the proposed ideas also apply to other regulations to a large extent.
(1)Grain Procurement Regulation and Rural Taxation Hypothesis:we argue that government grain procurement regulation is an important factor in explaining the problem of rural taxation.Everything else controlled,the higher the degree of government regulation enforcement (represented by government grain procurement as a percentage of total grain output),the higher the rural taxation rates as a percentage of rural (household and village level)net income,the higher the local government (village level)expenditure as a percentage of total local income,the more likely the local illegitimate fundraising and corruption..
(2)Grain Procurement Regulation and Factor Mobility Hypothesis:we argue that grain procurement regulation will limit the mobility of production factors such as land and labor.Everything else controlled,the higher the government grain procurement as a percentage of total grain output in a rural household,the less mobile its production factors,i.e.,the less likely its land will be leased out,the less likely its labors will migrate.
(3)Grain Procurement Regulation and Income Growth Hypothesis:we argue that by imposing heavy taxation burdens on peasants and not contributing much to local public good provision,and also by limiting factor mobility and deviating local and household level comparative advantages from being brought into full play,grain procurement regulation also have negative effects on rural income growth.Everything else controlled,the higher the government grain procurement as a ratio of total grain output in a rural village and household,the lower the income growth.A natural result is that more heavily regulated regions and households are further locked into agricultural production,and thus more vulnerable to government regulation on agricultural production and heavy rural taxation.
(4)With differentiating grain procurement regulation across regions (and also across households),regions (and households)that are more heavily regulated will have heavier rural taxation,more vulnerable to local bureaucracy expansion and serious corruption,lower factor mobility and thus lower income growth.The more heavily regulated regions (and households)will then be more locked in agricultural production,which will further lead to heavier rural taxation burden (as we argued in Part II)and even lower income growth.This constitutes a vicious cycle for these heavily regulated regions (and households),while the opposite happens to
less regulated regions(and households).Therefore,we can see that differentiating enforcement of government regulation will lead
to higher income disparity and more differentiating rural taxation burdens in rural areas.
More generally speaking,the regulation argument can be integrated into a more general framework proposed by Lin etc (Lin etc.,1999).According to them,many developing countries,including those socialist countries such as Russia,East Europe Countries and China,and even many Latin American countries and India that adopted import substitution strategy,were actually adopting an “overtaking”development strategy after World War II.In the authors’view,the traditional economic system in all those countries was aimed at a strategy of prioritizing capital-intensive heavy industries in a capital-scarce economy.Such a system was characterized by the trinity of a macro-policy environment of distorted prices for products and essential factors of production (e.g.trained personnel,funds,technologies,resources,etc.),highly centralized planned resource allocation and a micro-management mechanism in which firms had no decision-making power.China’s reform,being a gradual process similar to a Pareto improvement,starting with improving the micro-management mechanism by delegating powers to enterprises or regions and allowing them to retain a greater part of profits or fiscal revenue,was aimed at improving the incentive mechanism and micro efficiency.This reform weakened the inherent trinity in the original system.Later,the reform pushed forward towards improving the resource allocation system and the macro-policy environment,thus encroaching on the traditional development strategy.
However,as long as the government does not fully give up its overtaking development strategy that is incompatible with comparative advantage,it will result in the incompatibility of individual institutional arrangements,bringing about a host of difficulties in reform and development.
The government regulations on rural areas such as grain procurement and birth control are policy burdens in rural areas that can be viewed as residuals of the traditional development strategy.The reason is the following:the traditional overtaking development strategy requires the government to mobilize resources from rural areas for industrial development,while at the same time cannot provide sufficient employment opportunities.Therefore,the grain procurement system was established to provide food for industrial and urban development at prices below market level.At the same time,policies such as birth control and migration limitations must also be set up to reduce labor supply since the prioritized sectors are very capital-intensive.After the reforms from late 1970s,adjustments have been made in many respects,but the government is still unwilling or fear to give up its regulations on grain production and population growth,thus led to the many problems analyzed above.
The traditional literature on regulation usually emphasizes the role of regulations to promote interests of specific interest groups (Becker,1968;Stigler,1971).Impacts of government regulations on some sectors such as airlines,telecommunications were studied following this perspective.In development economics,literature on regulation usually focuses on the problem of impact of regulations on economic efficiency and development,i.e.,the problem of “helping hand”or “grabbing hand”.The more recent developments discuss the origin of regulation by studying the relationship between regulation and law,emphasizing the enforcement costs and the incompleteness of law (Shelifer,1998;Pistor and Xu,2002).However,up to now,no research has answered the following questions:in developing countries,why have so many governments adopted a whole set of complicated regulatory framework and regulate so extensively?If it is only aimed at predatory purpose by the state,is it necessary to regulate so extensively?A well--known fact is that as regulations extend to more aspects of economic and social life,bureaucracy will expand inevitably and conflicts among different interest groups will also intensify significantly.Therefore,extensive regulations might not be optimal for predatory purpose since costs entailed will be huge,as our example of government grain procurement and rural taxation signifies.
The development strategy framework provides a new workhouse for us to understand the formation,dynamics and impacts of regulatory regime in developing countries.The formation and dynamics of regulatory framework in developing countries is the outcome of formation and dynamics of the “traditional development strategy”.The regulatory framework in developing countries is established to serve the purpose of the overtaking development strategy in mobilizing resources to develop the prioritized sectors.Following this idea,we can trace the structure and evolution of regulatory framework in developing countries and understand more deeply the impacts of the regulatory framework on economic efficiency,income distribution and government behavior.At current stage,we can outline the prospects of further research along this direction.A more thorough analysis will be pursued later.This paper will focus on the problem of rural taxation and can be viewed as a first step in that direction.
*We are grateful to Minggao Shen,Yang Yao,Ying Qian and Shudong Hu for helpful comments.Mingxing Liu thanks the Ford Foundation and NBER for financial support.All errors are our own.
please quote the original source
if you are going to use this paper