Abstract
It no longer makes sense to talk of India without analyzing its infra-national diversity. Yet, this article not only argues for the need to build upon but also go beyond the subnational comparative analysis for India. I make three related points. 2 While scholars exploit the variation easily found at the provincial level in India, they must also take their subnational insights and generalize about India as a whole. Users of the subnational method must ask: How do the conclusions of subnational variation change or modify our understanding of India? Second, with economic liberalization and integration of markets within India, a focus on the subnational level makes forces that span across states and cities invisible. Is India becoming more integrated even as variation across its sub-state units is increasing? How can we understand both these phenomena in one analysis? 3 In order to understand both spatial differentiation and integration, we need to analyze diffusion and horizontal competition and processes of convergence across subnational units (Jenkins, 2000; Saez, 1999; Sinha, 2004). We can no longer look at policies at the subnational level without examining how policy innovation and e-governance spread across states. Last, the complexity of India’s internal variation makes us hesitate to do cross-regional and comparative studies. What can we learn from the subnational diversity of India, Brazil, China or Mexico studied comparatively? Can we compare such different countries especially when their internal variation makes easy national-wide descriptions suspect? While difficult to do, I would encourage more cross-regional and comparative studies that do not ignore the internal variation within India. 4 Overall, disaggregating the state to its lower levels may not be enough, and this article urges the need for a ʻscaling up framework’ as a complementary strategy to scaling down. Such a ʻscaling up framework’ must try to craft larger inferential statements about India, while keeping in mind its subnational diversity, the national or global context and the interstate experimentation relevant for the phenomenon under study.
We know that social and political processes unfold unevenly across countries as diverse as India, lending support for a spatially differentiated analysis. The multiple spatial variations affect colonial India, the workings of democracy, the changing processes of economic development and spatial patterns of economic liberalization, social policies and party systems. Methodologically, scholars have focused on states, cities, urban slums, party systems, SEZs and sub-regions to exploit greater variation within India (Auerbach, 2013; Bussell, 2012; Chandra, 2004; Harriss, 2003; Jenkins, Kennedy & Mukhopadhyay, 2014; Kale, 2014; Kohli, 1987; Singh, 2011; Sinha, 2005a; Varshney, 2002; Wilkinson, 2004; Yadav & Palshikar, 2003).
While truly comparative studies where more than one unit is analyzed comparatively and with some controls are still rare, many scholars focus on one province, and a large body of research has generated interesting case study research about specific provinces. Classic studies of a state include Sisson (1972), Robinson (1989), Brass (1974) and Weiner (1978), among many others. Thus, regional studies—scholars focusing on one region—have proliferated (Maharashtra—Birmal, Deshpande & Palshikar, 2009; Deshpande, 2006; Palshikar, 2014; Haryana—Jodhka, 2012, 2014; Tamil Nadu—Wyatt, 2013; Uttar Pradesh and Madhya Pradesh—Pai, 2002, 2010; Gujarat—Sinha, 2010; Kerala—Heller, 1999; Kerala—Singh, 2011). Rob Jenkins (2004) also edited a volume containing many innovative two-state comparisons that yielded many interesting insights about pairs of states. Interestingly, reflections on the methodological benefits of the subnational comparative method have added to our knowledge of different types of subnational studies (Pai, 2013; Snyder, 2001; Tillin, 2013; Vaid, 2013; Yadav & Palshikar, 2008). Subnational studies of varied stripes have cautioned against taking India as a homogenized entity too easily.
While this move to the subnational and lower levels of analysis has been indispensable and contributed to better knowledge about India, it deserves deeper scrutiny. This article argues for the need to take the strengths of ʻthe subnational turn’ in comparative politics of India and combine it with theory-building and cross-regional analysis. It may be time to use the strengths of this dominant methodological approach and build beyond the subnational comparative method. In this article, I make a case for scholars of India to use their sub-state insights to make conclusions about the national level, to conduct studies that deploy subnational insights towards cross-regional analysis (India and China, for example), to address issues about dynamic change across states as well as central-local interactions and model diffusion interactions and dynamics.
Towards Theory-building: What Do Subnational Findings Tell Us about India as a Whole?
The variation across its varied subnational units should tell us something about India as a whole. It is only when we take our subnational or city- or district-based insights and scale up that we will have a sense of the larger picture.
Theorizing about India is difficult, as it does not fit neatly into larger theories. Does India’s sub-national variety tell us something about the nature of the nation-state or state-nation? Sinha (2005a) argued that India is a segmented or divided state where different logics of development proceed at different paces making India more than a sum of its parts. Stepan, Linz and Yadav (2011) built a theory of India as a ʻstate-nation’ arguing that its different subnational entities give India a different state than a classic nation-state. They try to create a new typology of states generalizing from the internal diversity of India. Analysis of subnational variation should also yield more complex yet generalizable theories about India as a whole, apart from disaggregating the phenomena that we study.
Analysis across Levels: Diffusion and Horizontal Learning across Subnational Units
Cities and provinces are unlike nations in one important respect: Many of the dominant forces that affect them come from the federal structure or external influences (Peterson, 1981). Those who use the subnational method must be aware of this larger structural context. The conventional methodological justification for subnational analysis is that the national context can be held constant. However, with a changing national context, especially in terms of policymaking, the larger national context may be more dynamic than a temporally fixed level of analysis would suggest. For example, a national policy like the National Rural Employment Guarantee Scheme (NREGS), while implemented differentially across states, also changed the larger policy environment for all states. Moreover, the centre’s relations with sub-units may be heterogeneous, undermining the assumption that it can be a spatially fixed ʻthing’ in the background. 5 Methodologically, subnational units are subject to the same larger external influences creating a problem of autocorrelation (Lankina, 2012). At the very least, the relationship between the centre and its sub-units must be analyzed in a comparative subnational analysis.
States or cities are also linked with other similar or dissimilar units, and there is a huge amount of interaction across units especially in a federation like India. Indian states and cities are learning from each other and its time we focused attention on the processes of diffusion and learning underway within India. The international relations literature characterizes these in terms of diffusion, emulation and learning (Simmons, Dobbins & Garrett, 2008). Might we find similar processes shaping interstate interactions within India? I suggest that diffusion or horizontal competition across different units is usually ignored or even understated if we focus on the subnational level as an independent unit from other similar units. The proliferation of investment trips by chief ministers, single-window offices, digital governance policies, power reform across Indian states and new institutions of governance copied across states, all point to the power of diffusion and emulation at the subnational level. 6 More important, certain phenomena get overlooked in an attempt to scale down and compare across states. The creation of uniform sales taxes or the common goods and services tax is one such example.
In Sinha (2005a) I made a case not only for subnational comparison but also the necessity of looking at linkages across levels. How do its units bargain with each other? At that time, I found diverse strategies of bargaining in the field of large-scale industrial policy. More recently, the arena of fiscal policy and the bargaining over the common tax regime (common tax on goods and services) is the classic policy area where a bargaining model may work. There is a lot of politics in these interactions. A similar interaction over the Planning Commission and Finance Commission proceedings and output is worth analyzing. 7 Such bargaining and political interactions are unlikely to go away even with the replacement of the Planning Commission with NITA Aayog (National Institute of Transforming India). Another interesting research question is whether politicians and political parties learn from their counterparts in other states? An interesting case to analyze with this question in mind would be the emergence and success of the Telangana movement (2014), which may have learnt and used political strategies deployed by parties during the creation of three states: Jharkhand, Chhattisgarh and Uttaranchal (2000–2001). As integration and diffusion has proceeded apace in India, it makes sense to look at interaction and bargaining across levels and both theorize (model) and empirically test such interstate or intercity diffusion dynamics.
The Value of Subnational Cross-Regional Analysis: Exploring the Most Different Design
We need more puzzle-driven, cross-regional and comparative studies of India. India must not be seen as sui-generis or exceptional even as we give analytical weight to India’s internal complexities (Kohli, 2004; Sinha, 2005a, 2005b; Stepan et al., 2011). I wish to urge greater use of an unused comparative method, where India is compared with other countries, while keeping its internal diversity in mind. Some studies that adopt this methodological approach are as follows: Patrick Heller (2001—Kerala, South Africa, and Porto Alegre), Eric Jepsen (2006—India and Mexico), Sinha (2005b—India and China), Stepan et al. (2011—India, Sri Lanka, Ukraine and United States) and Lankina and Gertachew (2013—Russia and British India).
Let me summarize my own research on India and China, which sought to take the internal regional variation of two very different systems seriously. As I analyzed India’s and China’s reform experiences, I faced an interesting paradox. While a large theoretical literature called for analyzing the link between decentralization and economic reform outcomes, this literature posed India and China as contrasting countries. In contrast, I argued that in both systems, linkages across levels had been created that made economic reform politically sustainable. A focus on three distinct central–local mechanisms—of authority, institutions and personnel and the career incentives of politicians at both local and central levels—were compared across two very different systems. In this way, I sought to build a new theory to compare the distinct decentralization processes across India and China.
Admittedly, such cross-regional analysis is easier said than done. Those who conduct research at the subnational level invariably gather new data and create new measures and indicators, as very few truly comparable datasets exist about infra-national variations. This poses additional challenges to scholars and observers interested in comparative analysis as we need fine-grained subnational data and variables for more than one country (Lankina, 2012; Sinha, 2012). But it may be time to start having such conversations about how to gather such data and conduct such subnational-sensitive comparative analysis across nations. One way to address the issue is to start collaborative research projects with scholars of different regional expertise. I tried to do that in my article on India and Brazil (Olsen & Sinha, 2013).
India has been a natural home for the subnational comparative method. Now, we need to take the strengths of this method and engage in building of new typologies and theories, which are sensitive to its internal diversity. This should tie in well with comparative studies that go beyond India to compare other large nations such as China, Brazil, Russia, South Africa and even the US. With economic reforms, we also need to explore the common processes that are shaping many states in equal measure, creating diffusion and learning across sub-state boundaries. It’s time to build upon excellent research on sub-national variation within India and go beyond Indian exceptionalism.
Footnotes
Acknowledgements
Thanks to Divya Vaid for inviting me to write this article for Studies in Indian Politics. I would also like to thank Rina Kashyap and Adam Auerbach for their prompt and excellent comments on this article. I have learnt from many excellent subnational studies of India, many of which I could not cite given length constraints.