Abstract
In the largest territorial reorganization since the 1950s, when the modern state boundaries were demarcated, the Indian union government carved out three new states from three large north Indian states in November 2000. This was accompanied by discussions along political and sociological lines. Owing to a lack of data, the debates along economic lines were muted. Equipped with three and a half decades-long macro-panel data, we exploit this natural experiment, to investigate whether smaller states grow faster, in terms of per capita income. We construct five separate counterfactuals, using techniques such as synthetic control and elastic net regularization, for comparison. We do not find statistical evidence that support higher growth for the three erstwhile ‘combined' states, in the post-reorganization period, compared to their counterfactual. We further investigate the six states separately to see if the ‘new’ states grew at the expense of their ‘parent’ states, or vice versa. The state of Uttarakhand shows significantly higher growth compared to its counterfactual in the post-reorganization period. Two other smaller states (Bihar and Chhattisgarh) did grow faster than their counterfactual, but do not qualify for the statistical significance test. Three other states (Jharkhand, Madhya Pradesh, and Uttar Pradesh) also do not show a significant change in their growth trajectory. Overall, we find that the creation of smaller states may not be a panacea for their economic problems.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Data availability
Available on request.
Notes
Sociological studies have based their arguments on aspirations of the local population and their identities. Political scientists have argued from the political empowerment of the people as well as shifts in electoral politics. See among others, Majeed (2003), Mawdsley (2002), Prakash (2001), Tillin (2013). Also, see Rao (1973) which provides economic arguments for smaller states. Kumar (2010) summarizes arguments both in favor of and against the creation of smaller states.
It may be noted that there were a few other territorial reorganizations between 1956 and 2000 that gave birth to new states. For example, the bifurcation of the Bombay state in 1960 led to the creation of Gujarat and Maharashtra. In 1966, Haryana was carved out from Punjab. A few of the north-eastern states were created in the 1970s. Kudaisya (2014) revisits the 1956 reorganization of Indian states. Dholakia (2021) recently has made a six-decade long comparison (1961–2020) of Gujarat and Maharashtra, which were bifurcated in the mid-1960s.
In the speech made to the parliament (July 25, 2000), the then union home minister, Mr. L. K. Advani, emphasized that ‘administrative efficiency’ and ‘quality of governance’ were to be gained from the creation of these smaller states.
It may be noted that the power sharing between the union and the state governments is based on the ‘lists’ provided in the Indian constitution. All the subjects requiring legislation have been divided into Union, State, and Concurrent lists. The state governments have exclusive legislative powers over subjects falling in the state’s list. For the subjects falling under the concurrent list, the state governments have an important say, although the power of union government prevails in the case of any discord between the two. Most of the issues that do not require the involvement of other states come exclusively under the state list. Some of these include law and order, agriculture, local government, alcohol, land revenue and land records, public health, and sanitation. All these make state governments an important agency for welfare provision in India.
Difference-in-Discontinuities is a mixture of two methods: Regression Discontinuity Design (RDD) and Difference-in-Differences (DID). This is akin to applying the DID in the areas close to the border where a researcher expects that event could have distorted the outcome to be measured.
We are thankful to one of the reviewers for suggesting to include additional variables capturing governance and corruption. However, to the best of our knowledge, these are not readily available. This is even more unlikely for the period before reorganization. For instance, the earliest available Good Governance Index (GGI) is available for 2019. And one of the first systematic study about corruption, India corruption study, was conducted in 2005. It is important to note here that even if we find such state-level indices, even before 2000, they are unlikely to be available for the ‘smaller’ states before they became a separate state, which is essential for our purpose.
We source these annual data from various modules of Economic and Political Weekly Research Foundation’s India Time Series (EPWRF-ITS). These include: GSDP, bank credit, state budget, and electricity consumption. Literacy and IMR are from census publications. Road data are from BRS publication.
Also see Vaibhav & Ramaswamy (2022).
DID can also be specified in a regression framework. To make sure that parallel trends assumption holds, various control variables can be included in the regression setup (Khandker et al., 2009). However, it is often the case that the treatment also affects these time-varying control variables. This makes them ‘bad’ controls (Angrist & Pischke, 2008). Additionally, there exists the problem of serially correlated errors if one wishes to employ the panel characteristics of data, which happens to be in our case (Bertrand et al., 2004).
The acronym ADH is derived from the initial letters of the three co-authors Abadie, Diamond, and Hainmueller (Abadie et al., 2010). We use SCM and ADH interchangeably.
The per capita income of Bihar is the lowest among all major Indian states.
A fifth restriction, called exact balance, is also there. This implies that the treated group can be written as a linear combination of the control groups. This kind of approach is used in matching literature, where the treatment and control groups can be exactly matched. See Doudchenko & Imbens (2016) for details.
In appendix (Table 5), we summarize how five different methods can be achieved by applying one or more of these restrictions.
The estimation of the general structure, especially the Elastic Net (EN) regularization method, has its own set of computational problems, as discussed in Doudchenko & Imbens (2016).
The authors are thankful to Prof. Guido Imbens (Stanford University) for sharing the codes used in the study Doudchenko & Imbens (2016). The codes for Difference-in-Difference (DID), Elastic Net (EN), and Best Subset Selection (BSS) methods are written in R programming language. The codes for Synthetic Control Method (SCM) and constrained regression are written in MATLAB. The codes for the visual outputs—evolutions of outcome variables, the assignment of weight to different control units, and the standard errors—are written in MATLAB.
The standard error estimates and the weighting schemes, and their graphs, are available upon request.
We take the average values for these predictor variables in the pre-treatment period. This corresponds to 1981–99 for the combined state analysis and 1994–99 for smaller states analysis.
The Synthetic Control Method (SCM or ADH) assigned full weight to the state of Odisha to create ‘synthetic’ Bihar. So effectively, the state of Odisha is the control group in this case. This also explains the kink in the graph in 2003–04.
The Synthetic Control Method assigns positive weights to three states—Odisha, Rajasthan and West Bengal—to create synthetic state. The weights for Elastic Net method are, as with earlier case, more diverse.
Almost the entire weight is assigned to the state of Odisha.
The divergence in the behavior of BSS and elastic net in the case of Bihar may have been as a result of such an over-fitting.
Chhattisgarh also saw a political regime change in the state in the initial period, as with Bihar.
It may be kept in mind that the state that was carved out (Uttarakhand) formed a very small portion of combined UP. Therefore, this depressing or discouraging effect should not be fully attributed to the treatment.
Exploring the changes in state’s economic structure and underlying channel is beyond the scope of this paper. We are thankful to one of the reviewers for suggestions regarding exploring the tourism sector. We plot the shares of tourism sector in the state’s economy, the sector captured by Trade, Hotels, and Restaurant in the GSDP data (Appendix Fig. 13). For comparison, we also plot the same for industry and manufacturing sector. Prima facie, it appears that the change in shares of THR between 1999 and 2000 appears to be artificially inflated on account of change in the GSDP series.
We are thankful to one of the reviewers for suggestions regarding exploring ‘fiscal capacity’ in these reorganized states. We hope to pursue it in our future studies.
References
Abadie, A., Diamond, A., & Hainmueller, J. (2010). Synthetic control methods for comparative case studies: Estimating the effect of California’s tobacco control program. Journal of American Statistical Association, 105(490), 493–505.
Abadie, A., Diamond, A., & Hainmueller, J. (2011). Synth: An R package for synthetic control methods in comparative case studies. Journal of Statistical Software, 42(13), 1–16.
Abadie, A., Diamond, A., & Hainmueller, J. (2015). Comparative politics and the synthetic control method. American Journal of Political Science, 59(2), 495–510.
Abadie, A., & Gardeazabal, J. (2003). The economic costs of conflict: A case study of the Basque country. American Economic Review, 93(1), 113–132.
Acemoglu, D., & Robinson, J. A. (2012). Why nations fail: The origins of power, prosperity, and poverty. Crown Books.
Ahluwalia, M. S. (2000). Economic performance of states in post-reforms period. Economic & Political Weekly, 35(19), 1637–1648.
Alesina, A., & Spolaore, E. (2005). The size of nations. MIT Press.
Angrist, J. D., & Pischke, J. (2008). Mostly harmless econometrics: An Empiricist’s companion. Princeton University Press.
Asher, S., & Novosad, P. (2015) The Impacts of Local Control over Political Institutions: Evidence from State Splitting in India, Unpublished manuscript available at https://paulnovosad.com/pdf/asher-novosad-borders.pdf. Accessed 30 Aug 2023.
Bertrand, M., Duo, E., & Mullainathan, S. (2004). How much should we trust differences-in-differences estimates? Quarterly Journal of Economics, 119(1), 249–275.
Bhattacharjee, G. (2016). Special category states of India. Oxford University Press.
Bolton, P., Roland, G., & Spolaore, E. (1996). Economic theories of the break-up and integration of nations. European Economic Review, 3(40), 697–705.
Dhillon, A., Krishnan, P., Patnam, M., & Perroni, C. (2020). Secession with natural resources. The Economic Journal, 130(631), 2207–2248.
Dholakia, R. H. (2021). Comparing economic progress in Maharashtra and Gujarat, 1960–61 to 2019–20. Journal of Indian School of Political Economy., 33(1–3), 3–37.
Doudchenko, N., & Imbens, G. W. (2016). Balancing, Regression, Difference-in-Differences and Synthetic Control Methods: A Synthesis. Working Paper No. 22791, National Bureau of Economic Research.
Duflo, E., Glennerster, R., & Kremer, M. (2007). Using Randomization in Development Economics Research: A Toolkit. In T. Schultz, & J. Strauss (Eds.), Handbook of Development Economics (vol. 4, pp. 3895–3962).
Ghatak, M., & Roy, S. (2015). Two states: A comparative analysis of Gujarat and Bihar’s growth performance between 1981–2011. India Review, 14(1), 26–42.
GoI. (2001). National Human Development Report. Planning Commission, Government of India.
GoI. (2003). Report of the Committee on India Vision 2020 and Vision Documents of Some States. Planning Commission, Government of India. Academic Foundation.
GoI (2011). India Human Development Report. Planning Commission, Government of India.
Holland, P. W. (1986). Statistics and causal inference. Journal of American Statistical Association, 81(396), 945–960.
Kale, S., & Bhandari, L. (2010). Small states-large states. Analytique, 6(4), 3–16.
Khandker, S., Koolwal, G. B., & Samad, H. (2009). Handbook on impact evaluation: Quantitative methods and practices. World Bank.
Kudaisya, G. (2014). Reorganization of states in India- text and context. National Book Trust.
Kumar, A. (2010). Exploring the demand for new states. Economic & Political Weekly, 45(13), 15–18.
Majeed, A. (2003). The changing politics of states’ reorganization. Publius: the Journal of Federalism, 33(4), 83–98.
Mawdsley, E. (2002). Redrawing the body politic: Federalism, regionalism and the creation of new states in India. The Journal of Commonwealth & Comparative Politics, 40(3), 34–54.
Mazaheri, N., Al-Dahdah, E., Poundrik, S., & Chodavarapu, S. (2013). Leadership and institutional change in the public provision of transportation infrastructure: An analysis of India’s Bihar. Journal of Development Studies, 49(1), 19–35.
Mukherjee, A., & Mukherjee, A. (2015). Bihar: What went wrong? And what changed? In A. Panagariya & M. G. Rao (Eds.), The making of miracles in Indian States: Andhra Pradesh, Bihar, and Gujarat. Oxford University Press.
Prakash, A. (2001). Jharkhand: Politics of development and identity. Orient Blackswan.
Rao, C. H. H. (1973). Rationale for smaller states: The new imperatives. Seminar. Volume 164.
Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for generalized causal inference. Houghton Miffin.
Shenoy, A. (2018). Regional development through place-based policies: Evidence from a spatial discontinuity. Journal of Development Economics, 130(2018), 173–189.
Tillin, L. (2013). Remapping India: New states and their political origins. Oxford University Press.
Vaibhav, V., & Ramaswamy, K. V. (2022). Does the creation of smaller states lead to higher economic growth? Evidence from state reorganization in India. Working Paper. No. 2022/07. Indira Gandhi Institute of Development Research (IGIDR), Mumbai. http://www.igidr.ac.in/pdf/publication/WP-2022-007.pdf
Acknowledgements
This paper is based on the PhD Thesis submitted at Indira Gandhi Institute of Development Research (IGIDR), Mumbai by Vikash Vaibhav titled “State Reorganization and Development: Evidence from Indian Experience” in 2021. An earlier version was issued as IGIDR working paper (No. 2022/07) under the title “Does the creation of smaller states lead to higher economic growth? Evidence from state reorganization in India”. The authors gratefully acknowledge the comments received from R. Nagaraj, S. Chandrasekhar, and A. Ganesh Kumar at various stages of this study. The authors have benefitted immensely from in-depth methodological discussions with Sudha Narayanan, Sanjay Prasad, and Akhilesh Verma. The authors wish to thank Professor Hanumantha Rao and Professor R. Dholakia for their comments on the working paper version of this paper. The usual disclaimer applies.
Funding
The authors declare that no funds, grants, or other supports were received during the preparation of this manuscript.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Consent to participate
Not required.
Consent to publish
Not required.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
Compounded Annual Growth Rate (CAGR) is calculated based on the GSDP at constant prices (2004–05 prices). Combined Bihar is comprised of combined states of Bihar and Jharkhand. Similarly, Combined MP refers to combined states of Madhya Pradesh and Chhattisgarh. Combined UP refers to the combined states of Uttar Pradesh and Uttarakhand. ‘Reorganized’ states comprise the three states that were affected by reorganization (Combined Bihar, Combined MP, and Combined UP). ‘Poor’ states refer to four ‘poor’ Indian states: Assam, Odisha, Rajasthan, and West Bengal. ‘Northern’ states refers to three north Indian states: Punjab, Haryana, and Himachal Pradesh. ‘Western’ states refer to two West Indian states: Gujarat and Maharashtra. ‘Southern’ states comprise four south Indian states: Andhra Pradesh, Karnataka, Kerala, and Tamil Nadu. Finally, ‘Major Indian States’ is comprised of all 16 of these states.
Same as Table 1.
Author’s computations based on GSDP data.
Based on Doudchenko and Imbens (2016)
Share of tourism sector in Uttarakhand’s economy. The plot is based on constant prices data for the sectors considered, and for the aggregate GSDP. There are four constant price series we consider: 1993–94 series (for 1993–94 to 1998–99), 1999–00 series (for 1999–00 to 2003–04), 2004–05 series (for 2004–05 to 2010–11), and 2011–12 series (for 2011–12 onwards). Prima facie, it appears that the jumps/slumps in the tourism sector at 1999–00, and 2011–12, may be on account of change in series we use
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Vaibhav, V., Ramaswamy, K.V. State-size and economic growth: evidence from state reorganization in India. Ind. Econ. Rev. (2023). https://doi.org/10.1007/s41775-023-00199-4
Accepted:
Published:
DOI: https://doi.org/10.1007/s41775-023-00199-4