Rural Infrastructure in Odisha: An Inter-District Analysis

Rural Infrastructure in Odisha: An Inter-District Analysis Chittaranjan Nayak ABSTRACT The present study attempts to analyse inter-regional disparity in rural infrastructure in the state of Odisha. Three separate indices have been developed for different categories of rural infrastructure- physical, social and financial with help of the Principal Component Analysis before unifying them to a single index known as the Rural Infrastructure Index (RII). The study observes that there exists vertical inequality in the spread of different categories of infrastructure in the state. Disparity is the severest in the case of financial infrastructure. The study attributes underdevelopment of Kalahandi- BolangirKoraput (KBK) belt and some of districts of western- central Odisha to the underdevelopment of rural infrastructure. The analysis lauds the formation of special plans such as the KBK plan and formation of Western Odisha Council by the government. It calls for a time-bound delivery system and region-specific measures in place. Keywords: Rural infrastructure index, Principal component analysis, Odisha 1.0 Introduction Rural infrastructure involves the very socio-economic climate created by some special categories of rural facilities, which contribute to the development of the rural economy both by increasing productivity and by reducing unit cost in production. It is a critical supply-side factor, which plays an indispensable role in economic development. Its role may be more perceptible in agriculture and rural development. China‟s success with rural enterprise can be ascribed to „the provision of a minimum package of transport, telecommunications, and power at the village level‟ (World Development Report, 1994). _________________________ Dr. Chittaranjan Nayak, Lecturer, Department of Economics, Ravenshaw University, Cuttak, Odisha. 18 PRAGATI: Journal of Indian Economy Since economic reforms in 1991, strengthening infrastructure has become a new-fangled paradigm of India‟s development policy (Government of India 1996). In addition to the efforts of Government of India, states are also having their own policies in this regard. However, it is argued that state policies have regional prejudice (Schiff and Valdés, 1995; Sawant and Mhatre, 2000). An ESCAP study points: “Policy decisions to leave the allocation of resources to the market or to invest scarce resources in places with the best growth potential benefit some areas and regions over others” (p.8). Although many studies have attempted to analyse disparity in overall development indices, very few studies have addressed infrastructure disparity in India.1 Some studies have tried to examine linkage between infrastructure and economic development in India [Elhance and Lakshmanan 1988, Binswanger et al 1993, Gowda and Mamatha 1997, Datt and Ravallion 1998, Lall 1992 and 1999, Sahoo and Saxena 1999-00 and Ghosh and De 2004]. The above studies, however, have taken into account mostly urban infrastructure items, whereas issues concerning the provision of rural infrastructure services should be tackled in a different manner compared to those concerning urban infrastructure2. Only very few studies (Binswanger et al 1993, Bliven et al 1995, Bhatia 1999, Zhang et al 2001, Rao 2005) have analysed the progress and economic effects of rural infrastructure. Out of these studies, inter-state disparity in infrastructure is addressed by Bhatia (1999), which has attempted to build a composite index of rural infrastructure state-wise and examine the relationship between rural infrastructure development and growth in agriculture. Although innovative, it suffers from subjectivity and arbitrariness in selection of items and assignment of 1 2 Please see Bhatia (1999), Ghosh and De (2004) The population density in rural areas is much lower than urban areas. Due to sparse distribution of population, average cost of provision of basic goods is much more in rural areas in comparison to the urban areas. The average purchasing power of people in rural areas is also significantly lower than that of the urban areas. It is not expected that rural people can pay for the installation of basic goods. As a consequence, they may remain as a deprived lot for a longer period. It is observed that the average purchasing power of urban people is 180 per cent higher than average purchasing power of rural people. See India Rural Infrastructure Report published by National Council of Applied Economic Research (2004). Rural Infrastructure in Odisha- An Inter-District Analysis 19 weights (Nayak 2008). The present study takes into account these aspects in a different perspective. The present study attempts to analyse disparity in rural infrastructure in the state of Odisha. This eastern Indian state is considered to be one of the most backward states of India having a lot of potential. Measuring infrastructure development especially in a rural set-up involves several problems like selection of factors, assignment of weights, specifying time-dimensions, and problem of aggregation. These issues have been addressed in this paper. I have prepared composite indices of rural infrastructure for different categories of infrastructure. The paper develops as follows: Section II gives a brief description of the methodology and data base for the present analysis. Section III analyses the results of the study. Section IV concludes. 2.0 Methodology and Database This is a cross-section study based on secondary data for the year 2001. Data has been collected from different published sources like Census 2001(Orissa), Statistical Abstracts of Odisha, 2002; different issues of Economic Survey, Government of Odisha; Agricultural Census of Odisha, 2005; District Statistical Handbooks 2002 of all the districts of Odisha for the years. 2.1 Categorisation of rural infrastructure Rural infrastructure has been categorized into three broad categories, viz. Physical, Social and Financial infrastructures. Each category has again been subdivided to encompass different factors. The final selection of the items in each category has been made on the basis of fundamental reasoning and their significance on the predominant rural activity, i.e. agriculture. The details of finally selected items in each category of infrastructure have been presented in Table 1. 2.2 Normalisation Since the units of measurement of the selected factors are different, they give rise to the problems of aggregation. So the items have been normalised by deducting arithmetic means and dividing standard deviations to make them unit- 20 PRAGATI: Journal of Indian Economy free. Unit free measurement of different factors is essential for the development of a composite index. Table 1: Categorisation of Rural Infrastructure Categories of infrastructure Facilities Taken Variables taken Physical Irrigation Percentage of gross irrigated PGIA area to gross cropped area Social Electricity Percentage of rural households PHHELCT with electricity connection Transport Density of rural roads per thousand hectare of net sown RURDEN area PHHTELCN Rural literacy rate RURALIT Communication Education Health Housing Financial Abbreviation of variables Amenities Banking Credit Beds in rural allopathic BDHOSP hospitals per lakh of rural population PGDHOUSE Percentage of rural good houses PHHLATRN to rural total houses Percentage of rural households BNKSER availing banking services. Credit per operational holding AGCREDIT given by Agricultural Credit Co-operative Societies Marketing Marketing co-operative MKTGSOC societies per lakh of operational holdings Weighing method: Principal Component Analysis (PCA) 2.3 Preparation of Composite Indices We have prepared three composite indices such as Physical Infrastructure Index (PII), Social Infrastructure Index (SII), and Financial Infrastructure Index (FII) encompassing all the desired factors of infrastructure in the respective category and then combined them into a single composite index for rural infrastructure, known as Rural Infrastructure Index (RII). Rural Infrastructure in Odisha- An Inter-District Analysis 21 Contrary to the conventional methods of indexing by subjective weight assignment, the present study has employed the Principal Component Analysis (PCA) which is one of the approaches of factor analysis. Factor analysis attempts to identify the underlying variables or factors, which explain the pattern of correlations within a set of observed variables. Factor analysis is often used in data reduction by identifying a small number of factors, which explains most of the variance observed in a much larger number of manifest variables. In the PCA approach, the first principal component is that linear combination of weighted items, which explain the maximum variance across the observations at a point in time. Here the sole objective of the weighing mechanism is to explain the maximum variance for all individual indicators taken together across the districts at a point in time.The rationale of using the PCA is that it helps to reach an aggregate representation from various individual indicators. The infrastructure index is a linear combination of the unit free values of the individual factors such that Index i=Σ Wk Xki where Indexi = index of the ith district, Wk = weight of the kth factor and Xki = unit free value of the kth factor for the ith district. 3.0 Results and Discussion The present study has used both the Eigen value and the Bartlett Criterion for selection of principal components. It is observed that the first principal component explains around 54%, 57% and 49% of variances in the chosen normalised variables of physical, social and financial categories of infrastructure. The first principal component satisfies the Bartlett‟s criterion in all the three cases. Accordingly the indices are constructed as follows: 3.1 Physical Infrastructure Index (PII) PII = 0.684 PGIA + 0.957 PHHELCT + 0.877 PHHTELCN + (-) 0.0764 RURDEN It is observed that electricity (PHHELCT) has got the maximum weight followed by telecommunications (PHHTELCN) and irrigation (PGIA). However 22 PRAGATI: Journal of Indian Economy rural road density (RURDEN) has got negative weight. This is contrary to the expected lines. An analysis of the simple pair-wise correlation between the above factors reveals that RURDEN is either uncorrelated or slight negatively correlated to the other three factors (Table 2). This may be a reason for the unexpected sign as well as inconsequential weight of RURDEN. PGIA PHHELCT PHHTELCN RURDEN Table 2: Correlation Table PGIA PHHELCT PHHTELCN 1.000 0.535** 0.317 0.535** 1.000 0.840** 0.317 0.840** 1.000 -0.055 -0.093 0.044 RURDEN -0.055 -0.093 0.044 1.000 ** Significant at 1 per cent level of significance 3.2 Social Infrastructure Index (SII) SII= 0.843 RURALIT + (-) 0.734 BDHOSP + 0.442 PGDHOUSE + 0.909 PHHLATRN It is observable that PHHLATRN, a proxy measure of household amenities, assumes the highest weight in the social infrastructure category. As per the census, 2001, only about 7% of rural households in Odisha has got the facility of latrine of any kind. This indicates low level of progress in household amenities in the state. RURALIT and PGDHOUSE come in the second and third positions respectively in assignment of weights. RURALIT BDHOSP PGDHOUSE PHHLATRN Table 3: Correlation Table RURALIT BDHOSP PGDHOUSE 1.000 -0.532** 0.115 -0.532** 1.000 -0.067 0.115 -0.067 1.000 -0.695** -0.501** 0.458* PHHLATRN 0.695** -0.501** 0.458* 1.000 ** Significant at 1 per cent level, *significant at 5 per cent level However, BDHOSP, the representative of health infrastructure in the study, has surprisingly been assigned negative weight. It is evident from Table 3 that Rural Infrastructure in Odisha- An Inter-District Analysis 23 BDHOSP is negatively and significantly correlated to the other items of social infrastructure sans PGDHOUSE. It is quite possible since these factors are exogenously determined. Moreover, BDHOSP is inversely related to population. This means that, unlike the other factors, BDHOSP is likely to be lower in the populated coastal districts in comparison to the underdeveloped central and southern districts. This might be responsible for the negative weight of BDHOSP. 3.3 Financial Infrastructure Index (FII) FII = 0.04907 BNKSER + 0.859 AGCREDIT + 0.857 MKTGSOC AGCREDIT has received the highest weight, which is closely followed by MKTGSOC. BNKSER has got the lowest weight (0.049). The study observes that there is a relatively extensive network of agricultural credit co-operative societies in comparison to banking and marketing societies in Odisha. Low volume of transaction of banks in rural areas may be a cause of making its weight relatively lower. It is seen in Table 4 that there is significant correlation between AGCREDIT and MKTGSOC but BNKSER is almost uncorrelated to AGCREDIT and MKTGSOC. Table 4: Correlation BNKSER AGCREDIT MKTGSOC BNKSER 1.000 0.026 0.001 AGCREDIT 0.026 1.000 0.474** MKTGSOC 0.001 0.474** 1.000 ** Significant at 1 per cent level of significance, * significant at 5 per cent level 3.4 Overall Rural Infrastructure Index (RII) The Rural Infrastructure Index (RII) is the composite index of PII, SII and FII. This indexing has been done by using the PCA approach too. The similarities of the factors have been tested by the test of communalities. Since the first component explains around 83 per cent of total variance of the factors and the Eigen values of the other two components are less than unity, we have extracted the first component only. 24 PRAGATI: Journal of Indian Economy RII = 0.951 PII + 0.920 SII + 0.863 FII It is notable here that out of the three separate indices, PII has got the highest weight followed by SII and FII. All the three indices have been assigned positive weights in the making of RII. This is in line with expectations. 3.5 Disparity in Rural Infrastructure Development in Odisha The study makes use of coefficient of variation (CV) and the Gini coefficient as the criteria to understand the spread in different aspects of rural infrastructure. The districts have been ranked and categorised on the basis of RII. Disparity in Physical Infrastructure As per the level of development of rural physical infrastructure, the districts have been categorised as high physical infrastructure (High PI), medium physical infrastructure (Medium PI) and low physical infrastructure (Low PI) districts. All the thirty districts of the state have been vertically divided among the above three categories; comprising 10 districts each. It is noteworthy that Khurda, which entails the state capital Bhubaneswar, is way ahead of the other districts of the state in physical infrastructure (Table 5). The ratio of the most developed district (Khurda) and the worst performer district (Nawarangpur) in terms physical infrastructure development is as high as 5.73:1. Table 5. Physical Infrastructure Index: District-wise Division High PI Medium PI Low PI S.N District PII S.N District PII S.N District 1 Khurda 9.671 11 Baragarh 6.356 21 Gajapati 2 Cuttack 8.438 12 Sambalpur 6.183 22 Mayurbhanj 3 Puri 8.092 13 Dhenkanal 5.588 23 Koraput 4 Jajpur 7.810 14 Sonepur 5.330 24 Kalahandi 5 Bhadrak 7.299 15 Nayagarh 5.202 25 Malkangiri 6 Ganjam 7.257 16 Jharsugura 5.048 26 Bolangir 7 Jagatsingpur 7.011 17 Sundargarh 4.411 27 Nuapada 8 Kendrapara 6.818 18 Keonjhar 3.819 28 Kandhamala 9 Balasore 6.562 19 Boudh 3.783 29 Deogarh 10 Anugul 6.434 20 Rayagada 3.431 30 Nawarangpur PII 3.430 3.300 3.247 3.234 2.933 2.833 2.822 2.335 1.948 1.688 Rural Infrastructure in Odisha- An Inter-District Analysis 25 The study finds that most of the erstwhile undivided coastal districts of the north-eastern Odisha occupy relatively higher positions in PII in comparison to their south-western counterparts mostly inhabited by tribal people. Anugul is the only district from central Odisha, which could occupy a position, that too, the last position among the High PI districts category. Otherwise, the districts in the Middle PI group are located in the Central and in the Western Odisha. There is vertical inequality in physical infrastructure among the coastal, western and southern regions of the state. Apart from Sonepur and Rayagada, all other KBK (undivided Kalahandi, Bolangir and Koraput) districts are clubbed in the Low PI category. The tribal dominated Sundargarh district of the northern Odisha and Deogarh of the undivided Sambalpur district are also seen in the Low PI club. The analysis also explores development in the sub-items of physical infrastructure. Given the higher weights of rural electrification and rural teleconnectivity, it is well understood that the districts which fare well in these two infrastructure items are better performer in physical infrastructure development than the other districts of the State (Table 6). Amongst all the districts of the state, the KBK districts are the discriminated lot in physical infrastructure items, especially in rural electrification and teleconnectivity. Barring Malkangiri, almost all the KBK districts have poor irrigation infrastructure too. Thanks to the much-hyped drought situation and subsequent governmental activism, Kalahandi has got some respectability in irrigation factor but the situation of its sibling Nuapada, is one of the worst. Similar is the case of the undivided Bolangir district. While Sonepur has got the 5th position, Bolangir has got distant 28th position in the development of irrigation infrastructure. The analysis finds an unanticipated situation in the case of development rural roads. Most of the districts of coastal Odisha like Balasore, Bhadrak, Ganjam, Jagatsingpur and Kendrapara are coming in the lowest bracket in RURDEN, whereas some of the otherwise underdeveloped districts e.g. Gajapati, Kandhamala, Boudh and Deogarh have got ranks in the top ten districts in RURDEN. 26 PRAGATI: Journal of Indian Economy Table 6: District-Wise Ranking in Physical Infrastructure Items Rank in Districts Anugul Balasore Baragarh Bhadrak Bolangir Boudh Cuttack Deogarh Dhenkanal Gajapati Ganjam Jagatsingpur Jajpur Jharsugura Kalahandi Kandhamala Kendrapara Keonjhar Khurda Koraput Malkangiri Mayurbhanj Nawarangpur Nayagarh Nuapada Puri Rayagada Sambalpur Sonepur Sundargarh PGIA 24 15 9 4 28 12 3 23 16 19 10 11 6 29 13 27 2 21 8 20 7 17 30 22 26 1 18 14 5 25 RURDEN 7 30 24 28 26 4 18 5 10 2 25 23 20 11 16 3 21 29 9 27 19 22 14 6 17 1 12 8 15 13 PHHTELCN 2 10 13 8 23 19 6 30 12 20 3 9 11 4 27 24 16 18 1 22 29 26 28 15 25 7 21 5 17 14 PHHELCT 14 5 7 12 24 22 3 27 11 20 6 4 2 15 25 28 10 18 1 26 29 19 30 8 23 9 21 13 17 16 PII 10 9 11 5 26 19 2 29 13 21 6 7 4 16 24 28 8 18 1 23 25 22 30 15 27 3 20 12 14 17 Rural Infrastructure in Odisha- An Inter-District Analysis 27 This finding has two probable implications. One, the length of rural roads in these supposed underdeveloped districts is relatively lager than that of the coastal districts. Two, Net Sown Areas (NSAs) in the coastal districts are higher in comparison to the KBK districts. A careful analysis based on our observation of the study area points towards the later one. Most parts of the southern and some of the western regions of the state are densely covered by forests and hills, thereby the availability of NSA in these districts are lower in comparison to that of the coastal districts. So, physiographic factors come into the fore in explaining this inconsistency. Otherwise, the situation of rural roads is no way better in these underdeveloped districts. The above analysis, therefore, clearly indicates that there is both inter-regional and intra-regional diversity in the development of physical infrastructure in the state. Disparity in Social Infrastructure The categorisation of districts of the state in social infrastructure has been presented in Table 7. Here too we see the undivided coastal districts (except for Kendrapara) and Anugul are in High SI category whereas the undivided KBK districts with exceptions of Boudh, Bolangir and Sonepur come in Low SI category. Most of the districts of central Odisha are in the Medium SI category. The Western Odisha districts maintain their positions in the middle. The ratio of SII of the highest (Jajpur) and the lowest ranked districts (Malkangiri) is 5.006:1. Noticeably, there is a north-south divide in the social infrastructure development. The findings here are similar, with one or two exceptions, to the finding in case of physical infrastructure. Kendrapara of the coastal Odisha has marginally slipped to the medium SI category. Bolangir and Rayagada have swapped their places from low SI category to medium SI category. Some unusual revelations are brought about when we go for an item-wise analysis in social infrastructure. The districts, which are ranked higher in the health infrastructure, have got lower ranks in overall social infrastructure. This is due to the negative weight of BDHOSP, the selected parameter of health infrastructure. 28 PRAGATI: Journal of Indian Economy Table 7: Social Infrastructure Index: District-Wise Division Rank 1 High SI Dist Jaipur Rank 11 Medium SI Dist Kendrapara Low SI Dist Kalahandi SII 10.233 SII 8.090 Rank 21 SII 5.156 2 Cuttack 10.172 12 Dhenkanal 7.844 22 Mayurbhanj 4.993 3 Nayagarh 9.762 13 Jharsugura 7.275 23 Kandhamala 4.922 4 Khurda 9.713 14 Boudh 6.486 24 Gajapati 4.800 5 Ganjam 9.568 15 Baragarh 6.237 25 Koraput 4.687 6 7 Jagatsingp ur Puri 9.195 16 Keonjhar 5.780 26 Nawarangpur 4.432 9.077 17 Bolangir 5.721 27 Nuapada 4.149 8 Balasore 8.836 18 Sundargarh 5.658 28 Deogarh 3.898 9 Anugul 8.214 19 Sambalpur 5.423 29 Rayagada 3.546 10 Bhadrak 8.174 20 Sonepur 5.296 30 Malkangiri 2.044 It is also contrary to the general supposition that the coastal districts which are ahead of the central /south-western districts of Odisha have been ranked poorly in relation to BDHOSP (Table 8). It may be due to the simple reason that the coastal districts are densely populated and the availability of beds has not been according to the size of the population. A general impression is that the spread of urban health infrastructure is better in some of the coastal districts such as Cuttack, Khurda and Ganjam in comparison to that of the western and southern districts of the state. The impression from it might have been superimposed on rural areas of these districts too. But the present study proves it as an illusion eventually. The study observes that BDHOSP is lower in coastal districts than the KBK districts. The population factor explains this anomalous observation. The KBK districts and some of the districts of Central Odisha are sparsely populated. This makes BDHOSP higher in these districts in comparison to that of the coastal districts. Otherwise, BDHOSP is even below 15, which is perceived as the minimum requisite bed per lakh of rural population, in some of the KBK districts3. 3 Considering the incidence and severity of spread of diseases, the state government has also given emphasis to the KBK districts. However, the functioning and management of public health care system in these areas has always been a matter of controversy. Rural Infrastructure in Odisha- An Inter-District Analysis 29 It is clear from Table 8 that the districts, which are in better positions in PHHLATRN, are also better ranked in social infrastructure. Table 8: District-Wise Ranking in Social Infrastructure Items Districts Anugul Balasore Baragarh Bhadrak Bolangir Boudh Cuttack Deogarh Dhenkanal Gajapati Ganjam Jagatsingpur Jajpur Jharsugura Kalahandi Kandhamala Kendrapara Keonjhar Khurda Koraput Malkangiri Mayurbhanj Nawarangpur Nayagarh Nuapada Puri Rayagada Sambalpur Sonepur Sundargarh BDHOSP 15 27 10 28 17 30 26 5 18 13 20 23 29 9 16 2 21 14 22 12 1 11 19 24 7 25 4 3 8 6 RURALIT 12 4 13 5 21 18 7 19 10 26 17 1 8 11 24 22 3 16 6 30 29 23 27 9 2 2 28 14 15 20 Rank in PGDHOUSE 3 18 24 28 14 26 2 30 11 15 1 13 9 5 20 7 16 17 4 6 25 29 23 21 22 19 12 8 27 10 PHHLATRN 9 8 15 11 24 30 4 20 10 18 3 6 2 13 19 14 12 21 5 22 29 23 26 1 28 7 27 16 25 17 SII 9 8 15 10 17 14 2 28 12 24 5 6 1 13 21 23 11 16 4 25 30 22 26 3 27 7 29 19 20 18 30 PRAGATI: Journal of Indian Economy This is due to the highest weight assigned to this factor, which indicates that latrine facility is a positive and strong factor in social infrastructure in rural Odisha. The study takes this factor as a proxy measure of a number of household amenities like drinking water, kitchen, orchard etc. It is noticed that PHHLATRN is highly correlated to the other two factors, viz. PGDHOUSE and RURALIT. Once again it is seen that most of the coastal districts are in higher positions in relation to PHHLATRN whereas most of the KBK districts have got the bottom ranks. As regards PGDHOUSE, a coastal district Bhadrak has got one of the lowest positions (28th), whereas its sibling Balasore has got the 18th position. While Koraput has got the 6th position, the adjoining Malkangiri and Nawarangpur have got distant 25th and 23rd positions respectively. Similarly, while Bolangir has got the 14th position, its sister district Sonepur has got distant 27th position in PGDHOUSE. Here we notice an intra-regional disparity is more severe than inter-regional disparity. In RURALIT, however, the north-south divide is clearly visible. Almost all the KBK districts (except Nuapada) are in bottom positions, whereas almost all the coastal districts sans Ganjam have occupied top slots in RURALIT. Therefore, there exists inequality in the distribution of social infrastructure among and across the three major regions of the state. Disparity in Financial Infrastructure The districts too have been divided into three categories such as high financial infrastructure (High FI), medium financial infrastructure (Medium FI) and low financial infrastructure (Low FI) districts. The district-wise ranking has been presented in Table 9. Here we can see that Cuttack, Khurda, Ganjam and Nayagarh, which are ahead in PII and SII, are ahead in FII too. But some coastal districts such as Bhadrak, Jagatsingpur, Jajpur and Kendrapara have been placed in Medium FI districts category; Puri has been positioned in the Low FI districts category. On the other hand, tribal dominated Sonepur of the KBK districts; Sundargarh and Baragarh of North-Western Odisha have occupied positions in the High FI districts category. The performance of Nuapada and Kandhamala, two KBK districts of central Odisha in FI is visibly abysmal. Rural Infrastructure in Odisha- An Inter-District Analysis 31 Table 9: Financial Infrastructure Development Index: District-Wise Division High FI Districts FII Medium FI Districts FII 1 Cuttack 7.418 11 Sambalpur 3.425 21 Nawarangpur 2.127 2 Khurda 5.901 12 Bhadrak 3.334 22 Boudh 2.050 3 Ganjam 4.640 13 Jagatsingpur 3.248 23 Puri 1.977 4 Nayagarh 4.396 14 Jajpur 3.172 24 Rayagada 1.802 5 Sonepur 4.387 15 Jharsugura 2.827 25 Mayurbhanj 1.653 6 Sundargarh 4.129 16 Gajapati 2.660 26 Keonjhar 1.571 7 Baragarh 3.906 17 Kendrapara 2.549 27 Deogarh 1.425 8 Balasore 3.879 18 Bolangir 2.396 28 Kalahandi 1.364 9 Anugul 3.715 19 Koraput 2.330 29 Nuapada 0.841 Dhenkanal 3.487 20 Malkangiri 2.288 30 Kandhamala 0.741 Rank 10 Rank Rank Low FI Districts FII The distribution of financial infrastructure is not as asymmetric as it is in case of the other two categories of rural infrastructure. Though the southern districts including the KBK districts are far below their coastal and western counterparts in terms of development of financial infrastructure, yet the spread of FII between coastal and western Odisha is almost evenly balanced. It is seen that out of the three tribal dominated districts of North Odisha, Mayurbhanj and Keonjhar have been placed in the low FI category, whereas Sundargarh has got a place in the high FI category. It is noteworthy here that amongst the financial infrastructure items AGCREDIT has been assigned the highest weight, closely followed by MTKGSOC whereas BNKSER has been assigned the lowest weight (Table 10). As regards AGCREDIT, all the coastal districts sans Puri, Kendrapara and Jagatsingpur are well placed whereas all the KBK districts except Sonepur and Malkangiri have been ranked in lower stratum. The districts in the rolling uplands of Central Odisha have been ranked in middle stratum. The per centage of rural households availing banking services (BNKSER) is another factor in the making of the FII, albeit its low weight in overall FII. It is interesting to observe that the top three ranks have been shared by Anugul, Kandhamala and Nayagarh (Table 10), the three adjacent districts representing separate regions of the state whereas Kendrapara and Jajpur, the two 32 PRAGATI: Journal of Indian Economy neighbouring districts representing the same region have been placed in 6th and 26th positions respectively. Table 10: District-wise Ranking in Financial Infrastructure Items Rank in AGCREDIT BNKSER MKTGSOC FII District Anugul 12 1 5 9 Balasore 6 10 9 8 Baragarh 4 29 13 7 Bhadrak 8 18 16 12 Bolangir 25 30 7 18 Boudh 26 12 15 22 Cuttack 3 14 1 1 Deogarh 23 21 28 27 Dhenkanal 9 11 11 10 Gajapati 10 23 25 16 Ganjam 2 19 12 3 Jagatsingpur 17 4 8 13 Jajpur 7 26 22 14 Jharsugura 16 8 18 15 Kalahandi 29 24 19 28 Kandhamala 30 2 29 30 Kendrapara 19 6 10 17 Keonjhar 28 9 21 26 Khurda 5 16 2 2 Koraput 22 7 14 19 Malkangiri 14 25 26 20 Mayurbhanj 21 5 27 25 Nawarangpur 18 20 23 21 Nayagarh 1 3 20 4 Nuapada 27 28 30 29 Puri 24 22 17 23 Rayagada 20 13 24 24 Sambalpur 13 17 6 11 Sonepur 11 27 3 5 Sundargarh 15 15 4 6 Rural Infrastructure in Odisha- An Inter-District Analysis 33 Similar evidence is witnessed in case of the undivided Sambalpur district. While Jharsugura is in the 8th position, its adjoining district Deogarh has been placed in the 21st position. Here also disparity is more of intra-regional type than inter-regional. BNKSER is among the lowest in the undivided Bolangir and Kalahandi. Disparity in Overall Rural Infrastructure in Odisha The positions of different districts regarding rural infrastructure have been presented in Table 11 and Map 4. It is noticeable that there is vertical division among the three broad regions of the state in terms of rural infrastructure development. The coastal districts are ahead of their counterparts in the Southern and the Western Odisha. Table 11: Rural Infrastructure Development Index: District-wise Division High RII Medium RII Rank District RII 1 Cuttack 23.78 2 Khurda 3 Rank Low RII District RII Rank District RII 11 Kendrapara 16.12 21 Gajapati 9.974 23.22 12 Dhenkanal 15.53 22 Koraput 9.410 Ganjam 19.70 13 Baragarh 15.15 23 Mayurbhanj 9.158 4 Jajpur 19.57 14 Jharsugura 13.93 24 Kalahandi 8.996 5 Jagatsingpur 17.93 15 Sambalpur 13.82 25 Rayagada 8.081 6 Puri 17.75 16 Sonepur 13.72 26 Nawarangpur 7.518 7 Nayagarh 17.72 17 Sundargarh 12.96 27 Kandhamala 7.389 8 Balasore 17.71 18 Boudh 11.33 28 Nuapada 7.227 9 Bhadrak 17.33 19 Keonjhar 10.30 29 Deogarh 6.670 10 Anugul 16.88 20 Bolangir 10.02 30 Malkangiri 6.644 Out of the Central Odisha districts, only Anugul is in a relatively better position. This is due to better performance of the district in rural telecommunication, housing and banking infrastructure. The pace of industrialisation and the rural-urban linkage may perhaps be one of the factors for a relatively higher attainment of these basic infrastructures in Anugul district. Otherwise, all the other districts in the high RI category are from the undivided 34 PRAGATI: Journal of Indian Economy coastal districts of Odisha. Only Kendrapara has been clubbed in the medium RI category, that too in the highest position in the said category. Mostly districts from the Western Odisha, mainly from undivided Bolangir and Sambalpur, are in the medium RI category. Similarly, two northern districts Keonjhar and Sundargarh are in medium RI. However, another northern district Mayurbhanj is a low RI district. The same is the case of Deogarh. Though carved out of Sambalpur district, this district lags far behind the other districts of erstwhile Sambalpur. The vertical division between the sibling districts Gajapati and Ganjam is also evident. So in addition to inter-district variation, the study finds intra-district variations (within the erstwhile undivided districts) in rural infrastructure in the state. It is observed that all the present districts from undivided Koraput and Kalahandi districts have been categorised in the low RI group and almost all the districts in the low RI category are predominantly inhabited by the tribal people. Therefore, the governmental efforts to focus the Southern and Western Odisha through the KBK and the Western Odisha Council plans are steps in the right direction4. The scatteredness of different categories of infrastructure has been studied with help of Coefficient of Variation (CV) and Gini coefficient5 (Table 12). Table12: Descriptive Statistics Indices PII SII FII RII 4 Mean 5.07706 6.64598 2.98801 13.52124 Standard Deviation 2.15917 2.27126 1.47484 4.98443 Coefficient of variation 42.53 34.17 49.36 36.86 Gini Coefficient 0.238 0.191 0.264 0.206 The KBK plan is launched by the joint sponsorship of the central and the state governments with a view to focus on development of the three erstwhile undivided districts of Kalahandi, Bolangir and Koraput. 5 Gini coefficient (G) = 1=(1/n)-(2/n2 I) [I1 + 2 I2 +3 I3 +………………..+n In], where Ii , i=1,2,……….,n represent individual index in decreasing order of value, I is the mean value of the indexes and n is the number of districts, which is 30 in this study. Rural Infrastructure in Odisha- An Inter-District Analysis 35 We see that the CV of FII is the highest and it is the lowest for SII among all the indices. The same is the observation if we see the Gini coefficients. So, disparity is more severe in case of financial infrastructure. In the case of physical infrastructure more divergence is found in the irrigation variable PGIA. Relationship between Different Categories of Infrastructure It is normally expected that there is a positive correlation between different categories of infrastructure. The Karl Pearson‟s correlation coefficients have been calculated for this purpose (Table 13). It is found that the zero-order correlation coefficients between all the categories of rural infrastructure are highly significant. The physical infrastructure index is more associated with the social infrastructure index than the financial infrastructure index. Roughly speaking the set of base variables (PGIA, PHHELCT, PHHTELCN and RURDEN) of PII is more associated with the set of base variables (RURALIT, BDHOSP, PGDHOUSE and PHHLATRN) of SII than AGCREDIT, BNKSER and MKTGSOC of FII. Table 13: Pearson’s Correlation Coefficients PII SII FII PII 1.000 0.859** 0.731** SII 0.859** 1.000 0.648** FII 0.731** 0.648** 1.000 ** Significant at the 0.01 level (2-tailed) Underdevelopment in one aspect results in underdevelopment in another aspect of infrastructure. For example, if people have less access to physical infrastructure, their productive capacity is bound to be adversely affected. This may result in low accessibility to merit goods such as education and healthcare. Low attainment in the social field would reduce their bargaining power, by which they may remain alien to credit and marketing facilities. So, there always exists a cause-effect relationship among different categories of infrastructure. That is, however, beyond the scope of the present analysis. 36 PRAGATI: Journal of Indian Economy 4.0 Summary, Conclusion and recommendations The cross-section analysis in our framework develops composite indices for different categories of rural infrastructure viz., physical, social and financial infrastructure in the state of Odisha at district level. It is observed that the three indices are significantly correlated pair-wise, which indicates that underdevelopment in one aspect leads to underdevelopment in another aspect of infrastructure. These indices are further combined together to construct an index for the overall rural infrastructure. The analysis points towards existence of vertical inequality in the spread of different categories of infrastructure in the three principal regions of the state viz., Coastal, Southern and Western-Central Odisha. Disparity is more severe in case of financial infrastructure followed by physical and social infrastructure. The coastal region of the state is relatively better-off than the west-central and the southern regions of the state. The KBK districts, comprising districts mostly from Southern-Western Odisha, are in the lowest bracket of development in every aspect. This calls for proactive and participative role from concerned quarters so that all categories of infrastructure develop in synchrony and become adequate for rural development. Government efforts through the KBK plan and formation of the western Odisha Council are laudable steps in this direction. There is greater need for a timebound delivery system and certain region specific measures in place. The present study calls for revitalizing the existing rural infrastructure and evolution of a policy both at regional and national levels encompassing both the benefactors and the beneficiaries. References Bhatia, M.S (1999). Rural infrastructure and growth in agriculture. Economic and Political Weekly. 34 (13): A.43-A.48 Binswanger, Hans P., Khandker, Shahidur R. & Rosenzweig, Mark R. (1993). How infrastructure and financial institutions affect agricultural output and investment in India. Journal of Development Economics. 41: 337-66 Rural Infrastructure in Odisha- An Inter-District Analysis 37 Bliven, Neal, C.Ramasamy & Wanmali, Sudhir. (1995). Need for housing infrastructure. In Wanmali, S and C. Ramasamy (Ed). Developing Rural Infrastructure. Macmillan India Ltd. New Delhi. 28-51 Datt, G & Ravallion, M. (1998). Why have Some Indian States Done Better than Others at Reducing Rural Poverty?. Economica, 65 (1) Elhance, A.P. & Lakshamanan, T.R. (1988). Infrastructure-Production System Dynamics in National and Regional Systems: An Economic Study of the Indian Economy. Regional Science and Urban Economies. vol.18 ESCAP (2001). Reducing Disparities: Balanced Development of Urban and Rural Areas and Regions within the Countries of Asia and the Pacific. United Nations. New York. ST/ESCAP/2110. Ghosh, Buddhadeb & De, Prabir (2004). How Do Different Categories of Infrastructure Affect Development? Evidence from Indian States. Economic and Political Weekly. 39 (42): 4645-57. Government of India (1996). The India Infrastructure Report: Policy Implications for Growth and Welfare. Department of Economic Affairs. Ministry of Finance. Government of India. New Delhi. Lall, Somik V (1992). Infrastructure Investment and Economic Growth. Journal of Economic Perspective. 6 (4): 189-98. Lall, Somik V (1999). The Role of Public Infrastructure Investments in Regional Development. Economic and Political Weekly. 34 (12): 717-725. National Council of Applied Economic Research (2004). India Rural Infrastructure Report. New Delhi. Nayak, C.R.(2008). Physical Infrastructure and Land Productivity: A District Level Analysis of Rural Orissa, ICFAI Journal of Infrastructure. 6 (3): 7-21. Rao, C.H. Hanumantha (2005). What Constraints Agricultural Growth in India? Perspetives. The Economic Times, 27 September, p.7. 38 PRAGATI: Journal of Indian Economy Sahoo, Satyananda & Saxena, K K (1999-2000). Infrastructure and Economic Development: Some Empirical Evidence, The Indian Economic Journal. 47 (2):54-66 . Sawant, S.D. & Mhatre, Sandhya. (2000). Urban-rural Levels of Living in India: Trends in Disparity and Policy Implications. Indian Journal of Agricultural Economics. 55(2): 99-115. Schiff, Maurice & Valdés, Alberto. (1995). The Plundering of Agriculture in Developing Countries. Finance and Development. 32 (1): 44-47. World Development Report (1994). Infrastructure for Development. Oxford University Press. Zhang, Xiaobo & Fan, Shenggen. (2001). How Productive is Infrastructure? New Approach and Evidence from Rural India. EPTD Discussion Paper No.84. International Food Policy Research Institute. Washington DC., USA.