Rethinking the Developmental State Model: Divided Leviathan and Subnational

Comparisons in India
Author(s): Aseema Sinha
Source: Comparative Politics , Jul., 2003, Vol. 35, No. 4 (Jul., 2003), pp. 459-476
Published by: Comparative Politics, Ph.D. Programs in Political Science, City University
of New York

Stable URL: [Link]

REFERENCES
Linked references are available on JSTOR for this article:
[Link]
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@[Link].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
[Link]

Comparative Politics, Ph.D. Programs in Political Science, City University of New York is
collaborating with JSTOR to digitize, preserve and extend access to Comparative Politics

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
 Rethinking the Developmental State Model

Divided Leviathan and Subnational Comparisons in India

Aseema Sinha

For some time, the concept of the developmental state has been at the center of
debate in comparative political economy.1 It encompasses the view that state inter-
vention is necessary not merely to correct for market failures arising out of scarce
capital, externalities, and technological improvement but also to achieve long-term
capital accumulation and society-wide developmental goals. While the history of the
concept of the developmental state may lie with the economist Friedrich List, its
recent reincarnation was implicit in the writings of the post-World War II develop-
ment economists.2 It owes its current usage to Chalmers Johnson, who, in addition
to socialist and free market systems, posited a third category, the capitalist develop-
mental state. Johnson used this conceptual innovation to offer a revisionist account
of Japan's rise to power.3 The concept has survived the public choice onslaught
because of its empirical robustness in explaining the East Asian countries.4 The
World Bank for the first time in 1993 recognized that the state played a role in sus-
taining high growth, and subsequently its 1997 world development report was devot-
ed to analyzing the role of "the state in the changing world."5
India, regarded as a developmental failure, seems to challenge this emerging con-
sensus by confirming the public choice position that states negatively affect growth
by constraining private activity and generating rent-seeking. Amidst other relatively
prosperous Asian nations, negative images depict the Indian state as at once weak,
predatory, and interventionist. India has become the model of failure that helped
usher in the antistate, promarket Zeitgeist of the 1980s. Krueger's analysis of the
"rent-seeking state" originated in the Indian experience.6 The dirigiste developmen-
tal state in India outlined a powerful vision of state-led industrialization for the
whole nation-state in 1947, embodied in autonomous bureaucratic agencies, regula-
tions, and legislation. One high official in New Delhi is reported to have told a
friend: "If you want me to move the file faster, I am not sure I can help you; but if
you want me to stop a file I can do it immediately."7 The common and dominant
view was that the central state was responsible for India's slow growth rate.8 India's
unenviable reputation as a negative case continues to this day. While economic
reforms in 1991 ushered in a more promising scenario, their success is interpreted

459

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
Comparative Politics July 2003

through the old lenses. In a recent international conference, T. N. Srinivasan, an emi-

nent economist, lamented the persistence of a "wooden bureaucracy" even after the
onset of reforms.9 For others, the sweeping changes of the 1990s are valued posi-
tively only as long as they imply a drastic antidote to the erstwhile central state. The
Indian model of development seems to suggest that state failures are more crippling
than market failures, overturning a powerful postwar faith in state intervention.
Yet a set of intriguing puzzles, until now considered marginal to the debate over
India's developmental trajectory, confound and throw doubt on this picture. Despite
the powerful constraints of the central policy framework, some regional states within
India have managed to escape the adverse effects of the dirigiste regime. Diverse
regional states reveal very different patterns of industrial development and divergent
investment flows. Some (for example, Gujarat) attracted a very high share of indus-
trial investment, while others (for example, West Bengal) failed to capitalize on their
initial strengths as capital-rich states. Regional differences in outputs were accompa-
nied by institutional differences in the organization of investment. Some regional
states emphasized the role of the public sector; others encouraged synergistic public-
private coordination. These regional differences in outputs and institutions persisted
after radical changes were made in the regulatory system in 1991. Thus, if dirigisme
is the problem, as conventional opinion concludes, what explains the different sub-
national developmental pathways within India? Clearly, a similar central constraint
can not explain this variation. Did the successful regional states circumvent or miti-
gate the constraining effects of the central state in their regions? If they did, how
exactly did they do so?
These puzzling regional contrasts within a centralist nation-state challenge expec-
tations rooted in both statist and neoliberal accounts. Neoliberal theory argues that
the central state's policy framework is responsible for the laggard developmental
outcome in India. Regional differences, despite the presence of a central framework,
suggest that regional political elites circumvented or mitigated the effects of a con-
straining national environment, thereby reducing the analytical power of explana-
tions that attribute all investment flows to it. Statist theory focuses disproportionate-
ly on market failures; the high-performing states within India are not antimarket but
are market-enhancing. These doubts expose an important limitation in studies of the
political economy of development. Both neoliberals and statists have seen the state
as a unified actor that either succeeds or fails but in either case does so coherently on
a nationwide scale. Few studies have addressed the multilevel character of states;
consequently, there is no framework for systematically explaining variation within
developmental states, whether they are strong or weak. The political economy of
growth in most but especially in large countries must be understood with the help of
a multilevel, interactive model.10
This model posits that the policy framework of growth may not be centrally guid-
ed but is a joint product of central rules, provincial strategic choices, and subnational

460

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
 Aseema Sinha

institutional variation. The interaction between different levels of government is

consequential not only for political conflict and central-local relations, but also for
the implementation of central policy and the creation and sustenance of develop-
mental states.11 Cross-national analysis, which pays no attention to infranational dif-
ferences, compares noncommensurable units, throwing in doubt the validity and
generalizability of conclusions about developmental states. Yet subnational states are
not completely autonomous and thus can not be treated like nation-states. Rather,
located within a political hierarchy, they are only relatively autonomous. Thus, it is
necessary both to disaggregate the state and to reorganize its units in a multilevel
framework. The elements of this two-level framework are the incentive structure
created by (central) rules and regional political responses (regional strategies) to that
central framework, which may be a product of trade-offs in spatially different politi-
cal arenas (regional and national space). The strategic capacity of regional units to
implement their own developmental agendas also becomes crucial in this frame-
work.
Three comparatively chosen regional states within India help resolve the puzzle
of regional divergence in industrial investment flows and the institutional pattern of
public-private partnerships. The focus of the analysis is on the regulation of the
large-scale industrial sector from 1960 to 1991, the high point of the dirigiste state
and the period in which the currently observed regional patterns were set.12 An alter-
native multilevel framework that disaggregates states and points to the regional
political economy of governance challenges the national framework implicit in both
the market-friendly and the statist views in comparative political economy.13

The Puzzle of Subnational Variation within India

In India, despite the uniform national regulatory framework, the performance a

patterns of industrial development in various regional states differ markedly. Indi
provinces reveal a substantial diversity of development experiences in terms of o
comes but even more interestingly in terms of policies and institutions, as acros
nations. Investment data show that the divergent regional pattern had taken shape
1978 (see Table 1). The institutional organization of investment also shows a persi
tent regional pattern. In 1978, at the high point of the public sector developmen
strategy, 74 percent of investment in West Bengal's regional economy was constitu
ed by the public sector, 8 percent by the joint sector, and 17 percent by the pri
sector. Similarly, in Tamil Nadu the public sector absorbed 79 percent of investme
the private sector 15 percent, and the joint sector only 6 percent. Gujarat's owne
ship pattern was very different: 38 percent by the public sector, 30 percent by
joint sector, and 31 percent by the private sector.
What explains infranational variations within strong or failed developmen

461

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
Comparative Politics July 2003

Table 1 Per Capita Industrial Investment in Selected States in Rs.a

States 1978 1980 1986 1994

Investment Per Investment Per Investment Per Investment Per
Person (Rs.) Person (Rs.) Person (Rs.) Person (Rs.)
[Investment Index InvestmUent Index [Invemtment Index [Investment Index
(All India Per Capita (All India Per Capita (All India Per Capita (All India Per Capita
Investment =00)] Investment = 100)] Investment 100)] investment- I 00)]

All India 670 677 1234 9177

100oo] i100 (100] [1001

Gujarat 805 1142 2302 22776

[120] [1681 [186] [248]

West Bengal 428 469 927 4376

[631 1.69 [75] [47]

Tamil Nadu 517 503 629 7823

[77j [74] (501 185]

Maharashtra 779 617 1249 12385

[116] [911 [101] (134]

Bihar 684 704 410 2301

[102] [103] [331 [251

Kerala 361 367 504 5174

1.53J [54] [40] [561

Source: Computed from Center for Monitor

Come, 1978, vol. 1, 1-10; 1980, vol. 1, pp.l-1
"In current prices.

states? An important methodologica

ied. Should the focus be on cities (B
sus Surat), or regions (western ver
level) in investment flows is not
Moreover, the levers of political aut
the provincial level in India, and loc
power to affect investment patterns
aggregated to make analytical distinc
is the most appropriate unit of analy

462

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
 Aseema Sinha

Historical arguments suggest that initial political, economic, and social precondi-
tions were crucial for later developments. Regional patterns set during colonial times
had a powerful impact on later developments.14 Economists suggest that it is unfruit-
ful to compare advanced provinces (Maharashtra or Bengal) with backward ones
(Bihar or Orissa). Thus, a comparison of the industrial and economic structure
before the postulated divergence begins may be necessary to control for economic
variables. Selecting from the larger set of India's fifteen major regional states, I
chose those where many initial social and economic conditions were similar and
where developmental potential was clear. Following this reasoning, it was necessary
to choose those regional states that had experienced the economic and political
effects of colonial modernization directly (categorized as presidency states). Thus,
the choice of West Bengal and Tamil Nadu was clear (presidency areas). Two other
cases, Gujarat and Maharashtra (the southern part of Gujarat and most of
Maharashtra was part of the Bombay presidency), suggested themselves as plausible
cases for comparison.
Economically, the comparison of key industrial indicators by state on the eve of
independence shows the clear "dual dominance" of Bengal and Bombay in the
industrial field arising out of colonialism. 15 Historically, the colonial pattern of
industrial development predisposed the development of port towns, such as Bombay,
Calcutta, and Madras, which in turn worked as nuclei for the development of
Maharashtra (Bombay state until April 1960), West Bengal, and Tamil Nadu, respec-
tively.16 In contrast, the resource rich regions such as Bihar, Orissa, and Madhya
Pradesh lagged behind. The data on the spatial distribution of companies corroborate
the emergence of dual dominance, with Madras as the third most industrially
advanced state in the colonial period. In 1913-1914 the total number of companies
in the province of Bengal was 973 (35.4 percent), in Bombay, 613 (22.3 percent),
and in Madras, 427 (15.6 percent). In 1947, while Madras led in the number of reg-
istered factories, West Bengal was close behind; Bombay had the maximum produc-
tive capital, closely followed by Bengal, and Madras employed less than 10 percent
of productive capital. The three major provinces-Bombay (which included what
was to become Maharashtra and a large part of Gujarat in 1960), Bengal, and
Madras-accounted for 68 percent of total factory strength, and Bombay and Bengal
alone employed 62 percent of the total productive capital. Thus, these three presi-
dency areas were far ahead of other states in terms of industrial potential.
Indicators related to human and social capital-literacy rate and circulation of
newspapers--reveal that, while the literacy rate was similar across the three states in
1961 (30 percent in Gujarat, 29 percent in West Bengal, and 31 percent in Tamil
Nadu), the per capita circulation of newspapers was relatively unequal, with West
Bengal and Tamil Nadu having a significantly higher per capita circulation of news-
papers (1.5 and 2.7 times, respectively) than Gujarat in 1963.17 Thus, according to

463

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
Comparative Politics July 2003

human capital theory, Tamil Nadu and Bengal had the most conducive social condi-
tions for positive developmental outcomes, yet Gujarat's investment level outpaced
the other two. In 1951 all three states' share of urban population was similar:
Maharashtra's was 28 percent, Gujarat's 27 percent, West Bengal's 23 percent, and
Tamil Nadu's 24 percent. Kerala (13 percent) and Bihar (6 percent) ranked much
lower. Thus, on the eve of independence West Bengal, Bombay (Gujarat and
Maharashtra), and Madras (Tamil Nadu) were relatively well developed, urbanized,
and industrially advanced, thus generating the expectation that all four were poten-
tial developmental states.
I decided to focus on Gujarat rather than Maharashtra because Bombay city, the
capital of Maharashtra, contributes disproportionately to Maharashtra's develop-
ment.18 Given the dominance of Maharashtra on the industrial map of India,
Gujarat's trajectory of successful industrial management was not predetermined. It
could have become a subordinate hinterland, supplying labor and raw materials (oil
and lime, for example), to Maharashtra. Gujarat in fact became an independent cen-
ter of industrial activity, building autonomous sites of industrialization in southern,
central, and increasingly, in the 1980s, western Gujarat. The choice of economic
controls for the case selection implies that economic variables outlining either the
different initial conditions in the three cases or their different levels of industrial
potential can not fully explain the regionally divergent investment patterns across
them in the postindependence period.
It is important, however, to consider other plausible competing explanations.
Could the regional pattern have been a result of central bias in allocating discre-
tionary (public) investment to states ruled by an opposition party (for example, a
non-Congress party)? This explanation seems plausible. Opposition parties rule both
West Bengal and Tamil Nadu-the Left Front in West Bengal after 1977 and the
DMK and AIDMK in Tamil Nadu after 1967. Interestingly, this explanation does not
square with the evidence on discretionary central transfers to states.19 West Bengal
and Tamil Nadu, ruled by non-Congress party governments, consistently received
higher public sector allocations than Gujarat, which was ruled by the Congress party
from 1960 to 1990. The data on transfer of discretionary central resources to the
states shows that West Bengal, a state that receives a very low and declining share of
investment, receives a very high share of central public investment (see Table 2).
Thus, a focus on the central level can not explain regional divergence in investment
flows.
Another competing explanation is sociological. It suggests that the pattern of
regional class formation in West Bengal was not conducive to success in attracting
investment. This explanation focuses on Bengal to argue that the specific ethnic
character of the business class in Bengal was different from other states and the
dominant business community was "mercantile in spirit."20 West Bengal's economy

464

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
 Aseema Sinha

Table 2 State Share of Central Public Investment (Percent of All Indian Investment)

States Gujarat West Beugal Tamil

Nada
1960-1969 1.2 13.8 7.7

1970-1979 3.0 10.7 6.7

19801989 3.6 10.5 7.3

Source:Computed from, Bur

Industrial
and Commercial Und
Survey, Various Issus ew Delh

is dominated by the non-B

home province is Rajasthan
indigenous business classes. T
period when the difference
tive in the context of colonial
cult to argue that Marwaris
based in Bengal since the 18
have invested in industrial se
Moreover, there is no "indig
that attracts the largest inve
indigenous (Tamil) business c
This explanation, while theor
ment patterns, lacks empirica
tive framework that could e
Thus, economic factors, soc
tions do not fully explain th
tive analytical framework is

Developmental States: An A

Economic development is a
and time. The expanded role
of the development process
quences of growth give bure

465

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
Comparative Politics July 2003

ulation for political purposes. Moreover, because of the spatial embodiment of

development and state action, decisions about government's role in fueling econom-
ic growth can not be made independently from decisions about location.21 In large
states and in states with some differentiated regional traditions, regulatory policy
should be expected to become a tool in shaping regional economic prospects.
Furthermore, interests of regional and national politicians may vary.
Nevertheless, the theory of the state is held hostage to the received nation-centric
frame in the theory of economic development. Debates about the proper role of the
state in the economy usually assume a centralized national state. The framework out-
lined here suggests that it is necessary to pay more attention to the choices and
behaviors of regional rulers. Their (unintended) developmental impact on the nation-
al political economy emerges as a significant variable in analyzing economic policy-
making and developmental states. Two different yet interlinked vertical and horizon-
tal dimensions of developmental states need emphasis.
Recognition that the impetus for creating developmental states may also exist at
the subnational level puts regional elites and their strategic calculations at the center
of the analysis. Yet regional elites are not leaders of nation-states and are not com-
pletely free in their choices of policies and institutions; central rules and regional
constituents constrain their developmental visions and strategic calculations. Thus,
economic policy implementation is the product of intergovernmental interaction
among politicians located at different levels of the system. A fuller explanation of
developmental states requires a nested interactive framework that links higher level
constraints and regional democratic imperatives of regional rulers with the necessity
of enhancing regional investment flows.22 Vertical actions of subordinate rulers vis-
A-vis the central rulers are ultimately implemented in local or regional contexts;
thus, horizontal aspects arising out of the subnational political economy are equally
relevant. Subnational institutional capacities therefore become a key mediating vari-
able in explaining investment in the respective regions.23
Regional institutions may be defined as decision-making structures, bureaucratic
agencies, and formal and informal rules that are responsible for "translating" national
policy. In multilevel systems regional institutions may have an important impact on
investment flows and investors' decisions in three distinct ways: by affecting the cer-
tainty of transactions, by providing reliable information, and by enhancing the credibili-
ty of higher-level decisions and policies.24 Lobbying the central state on behalf of
investment proposals and ensuring that the procedures of the central bureaucracy work
will reduce the uncertainty faced by investors. Information is the key to investment and
economic growth. In systems where economic decisions are relatively free of govern-
mental control, information provision can be an important public good; some of this
information is a local public good. In systems where the (central) state has a monopoly
on information about economic decisions, local institutions may play a crucial role in
distributing that information and in generating new information. Thus, local informa-

466

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
 Aseema Sinha

tion services may substitute for informational rigidities at higher levels of the system in
both centralized and relatively decentralized systems. Table 3 table captures both the
vertical and horizontal dimensions of developmental states.

India

The central policy framework in India elaborated between 1947 and 1955 did not
realize its goals. Rather than organize a uniform dirigiste model of development, it
triggered the formulation of variable yet coherent long-term vertical strategies by
state-level incumbents toward the center. Some regional states sought to mitigate the
effects of the central state through bargaining and lobbying; others opposed the cen-
tral state. Investors responded to these subnational developmental models in differ-
ent ways, shaped more by regional institutional contexts than by the licensing frame-
work. The interactions between state and central governments fostered diverse mar-
ket governance patterns across India's provinces. They all emphasized the role of the
state, but the state interacted with the private sector in very different ways and
engaged the central state with variable effects.

Alternative Vertical Strategies Despite the common regulatory framework and

similarly statist policy repertoire, different regional state elites chose strikingly dif-
ferent vertical strategies towards the central state. In Gujarat political and bureau-

Table 3 Vertical and Horizontal Dimensions of Subnational States

VERTICAL MODEL=b BARGAINING CONFRONTATION

HORIZONTAL
CAPACITY

HIGH

A Vertically Integrated Isolationist but

Devetopmental State Productive State
(Gujarat)

Outcome: High Investment I(Outcome: Low-Mediumn

LOW

Statebut1Stat
Inconsisatenbt Integrated .,atiosist Non-Productive

(Taroi Nadr 1947-1967) (Wet Beagal)

Outcome: Low-Medium (Tnmil Ns 1967-1991)
Invtmnt AOutcome: Low Investment
. . . . . .. ... ........... . ----

467

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
Comparative Politics July 2003

cratic elites attempted to attract as much private sector investment as possible. They
monitored the output of the central regulatory regime and circumvented its con-
straining rules effectively. Vertical integration and high licensing flows were the
result. West Bengal's rulers, in contrast, followed a partisan confrontational strategy
that eschewed bargaining and monitoring. The level of confrontation with the center
was heightened. Subnational resistance and a partisan regionalism were the result. In
Tamil Nadu the party leaders oscillated between confrontation and lobbying. Their
strategies towards the central regime were contingent and varied over time. Table 4
summarizes the vertical model.
Gujarat's bureaucrats evolved a long-term and coherent infiltration strategy to
deal with the constraining rules of the regulatory system. Its essence is described
evocatively by a former Gujarat government bureaucrat. "Our motto was 'necessity
is the mother of invention'; the license-raj was a constraint and it led us to innovate
around it."25 Thus, while the actions of the Gujarat's bureaucracy embodied the clas-
sic developmental role of guiding markets, this market-guiding role of the subna-
tional state evolved through strategic interaction with the central state. Strategies of

Table 4 Three Strategic Models of Center-Regional Linkages

Bureuacratic Political Political

Liberalism Confrontation Protection
[Cooperationl [Confliet) [Mixed

Cases Gujantt West Bengal Tamil Nadu

Vertical Strategic Bureaucratic Confratational Protectionist
of Entrepreneurial Strategy Strategy
State-Level Strategy
tiltes Towards
the Center

1) Mode of Action Bargaining Confrontation and Alliance

towards the Center Eschew Bargaining formation or

Protest

(2) Channel Bureaucracy Media and Political Inter-Party &

of Access to the Arena Electoral
Center Alliances

(3) Timing of Long-t

Action Consistent Consistent Inconsistent
(varies with the
electoral cycle)

Type of Central- Pragmatic Ideological Pragmatic

Regional integration Regionalism Regionalism
Relatiouslip

468

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
 Aseema Sinha

monitoring and mitigation to counter the barriers to entry posed by central rules and
of circumvention to bypass rules that could not be mitigated evolved over time.
The first aspect of the strategy, "bureaucratic pressure," was evident in the con-
stant and careful monitoring of the regulatory process in New Delhi, a form of a
"industrial espionage."26 State agencies in Gujarat kept regular track of the number
of applications received by the central ministry, their current status, and the extent of
investment and employment envisaged by each project and their locations.27 Even
more important, the officials of Gujarat's government in New Delhi established con-
tacts with key ministry of industry officials to support Gujarat-specific investment
applications.28 These contacts involved the procurement of information on future
projects, allotment of licenses, checking of the status of licenses granted for Gujarat,
and follow-up to ensure that the applications for Gujarat were implemented effec-
tively and quickly. Obtaining crucial and usually informal information was key to
the process. As an example, information about prospective sectoral priorities of the
central ministries allowed Gujarat's government to propose "good" projects. These
efforts allowed Gujarat's industry department officials at licensing committee meet-
ings to build a coherent defense. Consequently, a large majority of Gujarat's invest-
ment applications was approved. Gujarat's state officials compensated for the rigidi-
ty of the central system in these diverse ways, mitigating its adverse effect on invest-
ment behavior.
Moreover, even when industrial capacities were frozen for the private sector, the
ministry of industry could approve public sector projects. The Gujarat Industrial
Investment Corporation (GIIC) applied for these licenses and after getting them
brought in private industrialists as joint sector partners.29 By 1978 30 percent of
investment in Gujarat's regional economy came from the joint sector, the highest of
any state. This tactic bypassed and circumvented the central rule of discouraging the
private sector. Thus, consistent monitoring, circumvention, and mitigation of the
central bureaucracy allowed Gujarat to ensure a high flow of investments to its
region.
In contrast, the West Bengal government did not monitor or mitigate the national
regulatory system. The West Bengal Administrative Reforms Committee, a govern-
mental body, commented as follows on West Bengal's office in the capital.

The state government [West Bengal] has to maintain a major presence in the nation's capital. It has
been the experience of other state governments that an effective liaison office in New Delhi facili-
tates dealings with the different ministries.... The condition of the liaison office the state govern-
ment [West Bengal] at present maintains in New Delhi is disappointing....Unfortunately, the quali-
ty of hospitality accorded by the State government to representatives of the Union government,
representatives of other State governments, members of parliament, legislators from other States as
well as foreign dignitaries leaves a great deal to be desired....We should be at par with what other
[regional state] Governments are capable of offering.30

469

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
Comparative Politics July 2003

The West Bengal political elite pursued an alternative strategy of conflict against
the center. After 1977, when the CPI(M)-led Left Front came to power, subnational-
ist opposition against the center was elevated to a long-term credible political strate-
gy. The sheer volume of official press output and rhetorical tactics on the question of
center-state relations in West Bengal is considerable. It ensured continuous public
attention to the question of central discrimination and shaped the political culture of
the state toward subnationalist resistance. One of the most famous rhetorical flour-
ishes was the public blood donation camp organized to raise money for a thermal
power project despite the absurdity of raising money in this way.
Approximately 60 percent of the documents released by the department of infor-
mation and cultural affairs of the government of West Bengal are on center-state
relations. Almost every public statement of key ministers on center-state relations is
published and circulated widely. A notable example is the publication of two vol-
umes that contain most of the chief minister's letters to the prime minster; the offi-
cial purpose of this publication was to "inform the public" about central discrimina-
tion.31 These letters embody the partisan confrontational strategy: their purpose was
to show the state government agitating against the center.
In addition, certain agencies were established to pursue these strategies. Political
and bureaucratic organizations such as the information and cultural affairs depart-
ment, Ganashakti, the Bengali newspaper of the CPI(M), and People 's Democracy,
another CPI(M) newspaper, as well as the support given to various cultural and edu-
cational agencies formed the centerpiece of a strategic model of political confronta-
tion. In contrast to the public and cultural channels utilized by West Bengal's rulers,
Gujarat's leaders preferred to lobby the central government through bureaucratic
channels. Both strategies, although contradictory, were successful on their own
terms because they were credibly sustained by the state's political elites and the cor-
responding institutional innovation in both states.
Regional actors in Tamil Nadu adopted defensive and protectionist postures to
protect a culturally specific populist strategy. They thus pursued inconsistent strate-
gies toward the central government that varied over time. The DMK and later
AIADMK (regional parties ruling Tamil Nadu after 1967) oscillated between con-
frontation and bargaining. These parties' roots lay in a strong regional populism and
deployed its antinorth and anti-Hindi rhetoric to oppose the hegemony of the central
state from 1967 to the late 1970s. In 1969 Tamil Nadu's government commissioned
a report on center-state relations, the Rajamannar Report, which was extremely criti-
cal of the central government. An industrialist reported: "We are missing out a lot by
not being close to Delhi. In the last ten years a lot of money changed hands in terms
of agency commissions, turnkey projects and World Bank aided projects [through
the central government]. We are not daring enough to be part of the scene."32 At this
time, no attempt was made to mitigate or monitor the central regulatory agencies.

470

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
 Aseema Sinha

Suresh Krishna, CEO of a Madras-based automobile company called TVS, felt very
strongly that Tamil Nadu's government did not lobby enough with the center in the
1970s and the 1980s. In 1984 Krishna suggested to Tamil Nadu's government: "As a
first step, the Tamil Nadu government should establish close liaison with the Center
so that its existing strengths can be capitalized on for locating any new manufactur-
ing unit...there can be no justification for the existing vehicle manufacturing in
Tamil Nadu having to look elsewhere for better terms and opportunities due to lack
of initiative by the home State."33 However, alliance and electoral exigencies in the
1980s forced the government of M. G. Ramchandran, chief minister of Tamil Nadu
from 1977 to 1988, to seek favors from the center. Prime Minister Indira Gandhi
promised key industrial projects in the 1980s in return for regional electoral support.
Thus, regional party politicians in Tamil Nadu followed an inconsistent ad hoc strat-
egy that was driven by their cultural and electoral compulsions. The flow of invest-
ment to Tamil Nadu thus oscillated from low to medium.

Horizontal Institutional Analysis Investment flows are affected only when the
vertical strategies towards the center are complemented by institutions that provide
regionally specific enabling environments. Clearly, the investors could not rely on
the central government for speedy implementation or reduction of their contracting
risks. Some regional institutions compensated for this weakness in central institu-
tional design; in doing so, they enhanced the credibility of state-level institutions
that enabled greater investment than would have been expected under a centralized
regime. Regional institutions exhibited wide variation in institutions, rules, and
styles. This variation was consequential for investment. One aspect is emphasized
here: information provision.34
Information provision by the regional states affected the uncertainty and credibil-
ity of implementation, compensating for informational rigidities in the system. The
comparative analysis of information services revealed that Gujarat, by ad hoc exper-
imentation, designed institutional mechanisms to collect industrial information and
then disseminated it to industrial entrepreneurs. In its effort to ensure investment
flow from Bombay, Calcutta, and East Africa in the 1960s and 1970s, it generated
databases of the potential entrepreneurs and provided one rationale for the creation
of a specialized body, iNDEXTb (Industrial Extension Bureau), to collect data and
information. Even more significant, in an explicit effort to develop sectoral leader-
ship, for example, in chemicals in the 1960s and in electronics in the 1990s,
iNDEXTb and other industrial agencies collected industry-related information about
many sectors.35 In addition, information about government rules at both the central
and state levels was collated in one place.
West Bengal and Tamil Nadu did not create such an agency.36 They had a statisti-
cal department, which collated industrial data, but its role was passive. None of the

471

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
Comparative Politics July 2003

information was made available to investors in West Bengal and Tamil Nadu. The
industry departments in the two provinces provided information regarding various
incentives, but it was dispersed and was not coordinated and collated. Information
was fragmented and not easily available.37 Sectoral or regional studies were not con-
ducted by the industry departments in the two states.
A comparison of state-level promotional agencies shows that the vertical strategies
toward the central government need to be complemented with state action towards the
private sector. This action ensures higher investment flow as well as a higher implemen-
tation of investment intentions. In those states where the institutions lacked horizontal
capacity, investment flows were correspondingly weaker. Thus, subnational develop-
mental states within India combine both vertical and horizontal dimensions with signifi-
cant consequences for regional investment patterns and national regulatory policy.

Concluding Observations

Are the conclusions of this analysis specific to India, or can they be generalized to
policymaking in other states and to larger theoretical debates? India provides three
significant contributions for broader theoretical debates with implications for other
cases: recognition of the need to address the problem of the state comparatively, a
nuanced view of the state versus market dichotomy in political economy and studies
of neoliberalism, and a methodological critique of cross-national studies.
Studies of the developmental state either argue for its importance or suggest that
states' solutions generate their own problems. This analysis, by contrast, disaggre-
gates the state spatially, showing how variation within a state and interaction across
governmental levels may affect investment flows and contribute to national develop-
mental failure. An alternative framework can incorporate size and geography in
explanations of developmental states. This framework consists of two, vertical and
horizontal, models, linked to each other in a two-level interaction. The dual focus on
regional elites' strategic choices toward the central rules and horizontal institutional
variation shapes the nature of national-level regulation and regional investment
flows. This linkage of different spatial arenas in analyzing national-level regulation
must modify insights generated by comparative political economic debates that rely
exclusively on national governance models.38 Studies of substantial subnational
variation in the pursuit of economic policy in Brazil, Mexico, and China strengthen
the point that developmental states exist at different levels within a nation-state.39
Subnational variation has a vertical intergovernmental dimension that has not yet
been analyzed explicitly.40 Attempts at subnational re-regulation may be engendered
by the policy imperatives of the central rulers. A multilevel framework is needed to
theorize linkages and interactions across levels. In large countries with some degree
of multitiered authority, increasing or fine-tuning central or subnational state capaci-

472

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
 Aseema Sinha

ty, while important, may not be enough. The type and extent of conflict and competi-
tion within the system will also affect subnational initiatives and the output of
nation-states. Hence the developmental state needs a theory of multitiered decision
making within the nation-state.
Spatial or central-local dynamics affect national developmental patterns. The nation-
al state is not exhausted by the actions of its central rulers. It is a complex aggregation
of lower level states and local elites. The provincial states are poised between regional
society and the larger state system. Depending upon their strategic capacity, they can
either create an independent developmental state, subtly wresting power away from the
central state, or become dependent states. Analysis of the full, multilevel complexity of
the regional actors' relationship to the nation-state and the external world is necessary to
explain how the nation-state system is both made and remade.
Differences across Indian states do not derive from state or market based action
but rather from different types of state-led strategies that combine different organiza-
tional forms, for example, the joint sector, to manage development. The varied
responses within India suggest that developmental states come in different types and
embody different microinstitutional mixes of public and private principles. Cross-
national comparisons may fail to illuminate such crucial fine-grained variables.
These findings can be applied to similar evidence of mixed organizational forms in
China's provinces and in eastern Europe.41
An important priority for future research involves analyzing subnational and
intergovemmental interactions in less dirigiste regimes and in neoliberal policy tran-
sitions. Withdrawal of central state regulations over markets may provide greater,
not lesser, opportunities for provincial governments to re-regulate economic poli-
cies; thus, liberalization has enhanced local state capacity in many contexts.42 In
addition, vertical interactions with the central state or international actors (such as
the World Bank and multinational corporations) may also remain necessary. The
regional states in India, for example, continue to lobby the central government for
World Bank projects, international loans, foreign investment, and central transfers
even after liberalization.43 A multilevel interactive model implies that vertical strate-
gies and institutional variation at the subnational level are as relevant for dirigiste
developmentalism as for liberalization.
Methodologically, this argument offers important lessons for traditional cross-
national analysis.44 Comparative cross-national studies of economic performance are
troubled by too few cases chasing too many variables. Infranational institutional analy-
sis makes it possible to develop finer analytical categories that control for nationwide
variables and emphasize subnational microinstitutional variables. Comparative political
economy should focus on new units of analysis and a multilevel framework.
Thus, regional differences and the politico-economic conflicts arising out of them
may crucially shape the nature and the output of the national political economy.
Regional differences not only influence identity formation but also the formation of

473

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
Comparative Politics July 2003

economies. The activities of the constituent parts can help explain the whole.
Unraveling the riddle of regional developmental states can illuminate not only the
political economy of industrialization, but also the nature of the state in general. It is
time to reorient comparative studies from the nation-state to a multilevel regional
political economy of governance.

NOTES

The author would like to thank Anindya Saha, Aili Tripp, and anonymous reviewers for extrem
able comments on this article. I also wish to thank Mark Beissinger, Richard Bensel, John E
Gent, Edward Friedman, Ron Herring, Mary Katzenstein, Peter Katzenstein, Melanie M
Chandra Mohan, Hector Schamis, Chris Way, and David Weimer for comments on the larger
which it is a part. I am thankful to the American Institute of Indian Studies and the Institute for
of World Politics for financial support. The interviews cited were conducted by the author.
1. The word "state" has multiple usages in this article. State with a small s refers to the ge
cept of the state. Indian provinces are also called states and are referred to as provinces or regi
The federal government is usually referred to as the central state or center.
2. For the idea of a big push, a classic statist idea, see P. N. Rosenstein-Rodan, "Prob
Industrialization of Eastern and Southeastern Europe," Economic Journal, 53 (June-Septem
202-11; Allyn Young, "Increasing Returns and Economic Progress," Economic Journal, 38 (
1928), 527-42.
3. Chalmers Johnson, MITI and the Japanese Miracle: The Growth of Industrial Policy, 1
(Stanford: Stanford University Press, 1982). Also see Meredith Woo-Cumings, ed., The Dev
State (Ithaca: Cornell University Press, 1999).
4. See Alice Amsden, Asia 's Next Giant: South Korea and Late Industrialization (New York
University Press, 1989); Johnson; and Robert Wade, Governing the Market: Economic Theor
Role of Government in East Asia (Princeton: Princeton University Press, 1990).
5. World Bank, The East Asian Miracle: Economic Growth and Public Policy (Washing
Oxford University Press, 1993).
6. Anne O Krueger, "The Political Economy of the Rent Seeking Society," American
Review 64 (June 1974), 291-303.
7. Pranab Bardhan, quoted in Dani Rodrik, "The Paradoxes of the Successful State," E
Economic Review, 41 (April 1997), 434.
8. Between 1960 and 1986 India's aggregate performance was one of slow growth, 3.5 per
annum (1.8 percent per capita per annum). The perception that the central state is responsible f
developmental failure cuts across contending approaches. See Pranab Bardhan, The Political E
Development (Oxford: Basil Blackwell, 1984); Lloyd I. Rudolph and Suzanne Hoeber Rudo
Pursuit ofLakshmi (Chicago: University of Chicago Press, 1987); and Isher Ahluwalia, Industr
in India: Stagnation since the Mid-Sixties (Delhi: Oxford University Press, 1985).
9. Aziz Hanifa, "Bureaucracy Is a Major Threat to India's Economic Ambitions," India Abr
14, 2001, p, 38.
10. See E. A. G. Robinson, ed., Economic Consequences of the Size of Nations (New
Martin's Press, 1960); John Lewis, India 's Political Economy. Governance and Reform (Delh
University Press, 1995); Richard E. Feinberg et al., eds., Economic Reform in Three Gia
Brunswick: Transaction Books, 1990).
11. This framework is applicable to the politics of development in any large state (for

474

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
 Aseema Sinha

Russia, China, and Brazil) but also to small states with significant territorial differentiation (for example,
Italy).
12. Systematic regional variation in industrial development under a state-led policy regime challenges
received economic and political models of dirigiste societies for two reasons. First, explicit policy instru-
ments designed to prevent regional inequalities were present. Second, industry was a central subject, and
regional differences are counterintuitive in this policy domain. Although the focus of this article is on the
dirigiste period of India's developmental trajectory (1955-1991), its conclusions are relevant for less
directive (market-based) systems and periods (post-1991 India). Regional choices and local institutional
variation refract such policy transitions as well. For reasons of space it is not possible to elaborate on this
point. See Aseema Sinha, "Ideas, Interests and Institutions in Policy Change in India: A Comparison of
West Bengal and Gujarat" (unpublished).
13. Richard Locke, Remaking the Italian Economy (Ithaca: Cornell University Press, 1995), in a simi-
lar vein challenges the national model framework of comparative political economy. Also see Gary
Herrigel, Industrial Constructions: The Sources of German Industrial Power (Cambridge: Cambridge
University Press, 1996).
14. Amiya Bagchi, "Reflections on Patterns of Regional Growth during British Rule," Bengal Past
and Present (January-June 1976), 247-89.
15. See Claude Markovits, "Bombay as a Business Centre in the Colonial Period," in Sujata Patel and
Alice Thorner, eds., Bombay: Metaphorfor Modern India (Bombay: Oxford University Press, 1996).
16. Amiya K. Bagchi, Private Investment in India 1900-1939 (Cambridge: Cambridge University
Press, 1972); T. R. Sharma, Location oflndustries in India (Bombay: Hind Kitab Limited, 1946).
17. Computed from Registrar of Newspapers for India, Press in India, 1965 (Delhi: Government of
India, 1965).
18. "In fact, Maharashtra excluding this [Bombay city] region, is industrially as backward as Uttar
Pradesh, Bihar or Rajasthan." A. R. Kamat, "Politico-Economic Developments in Maharashtra: A Review
of Post-Independence Period," Economic and Political Weekly, 15 (October 1980), 1671.
19. Discretionary transfers refer to central public sector allocation. Other central transfers such as the
Finance Commission transfers were not discretionary but were governed by statutory rules.
20. Debdas Banerjee and Anjan Ghosh, "Indian Planning and Regional Disparity in Growth," in A. K.
Bagchi, ed., Economy, Society and Polity: Essays in the Political Economy of Indian Planning (Calcutta:
Oxford University Press, 1988).
21. Douglas Arnold, "Legislators, Bureaucrats and Locational Decisions," in Mathew McCubbins and
Terry Sullivan, eds., Congress: Structure and Policy (Cambridge: Cambridge University Press, 1988), p.
524, makes a similar point about regulatory policy in the United States.
22. George Tsebelis, Nested Games: Rational Choice in Comparative Politics (Berkeley: University
of California Press, 1990). Also see Peter Evans et al., eds., Double-Edged Diplomacy: International
Bargaining and Domestic Politics (Berkeley: University of California Press, 1993).
23. See Elizabeth Remick, "Building Local States," Comparative Politics, 34 (July 2002).
24. See R. Fernandez and Dani Rodrik, "Resistance to Reform: Status Quo Bias in the Presence of
Individual-Specific Uncertainty," in Federico Sturzenegger and Mariano Tommasi, eds., The Political
Economy of Reform (Cambridge, Mass: MIT Press, 1998); Keith Krehbiel, Information and Legislative
Organization (Ann Arbor: University of Michigan Press, 1991); and Douglass North, "Institutions and
Credible Commitment," Journal ofInstitutional and Theoretical Economics, 149 (March 1993), 11-23.
25. Interview with N. Vittal, Gujarat government official, December 1997. License-raj refers to an
elaborate central regulatory system whereby the central government had the legal power to grant approval
of investment applications (called licenses), determine location, circumscribe technological choices, and
shape ownership patterns.
26. Interview with a retired Industry Secretary of Gujarat, Chennai, September 16, 1997.
27. Interview with an Industry Department official, April 29, 1997.

475

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]
Comparative Politics July 2003

28. Each regional government has an office in New Delhi; some regularly lobby the central govern-
ment.

29. Interview with a Gujarat Industrial Investment Corporation official, New Delhi, December 1997.
30. Government of West Bengal, Report of the Administrative Reforms Committee (Calcutta:
Government of West Bengal, 1983), pp. 25-26.
31. Jyoti Basu, Chief Minister s Letters to the Central Government (Calcutta: Government of West
Bengal, 1981, 1982).
32. Interview with a Madras-based industrialist, Business India, Jan. 23-Feb. 5, 1989, p. 56.
33. "Special Feature on Tamil Nadu," The Hindu, June 21, 1984, p. 29. Suresh Krishna reported in an
interview that in the 1970s and 1980s the state government did not lobby the center for central or private
investment. Interview with Suresh Krishna, Chennai, September 1997.
34. For reasons of space, analysis is limited to information provision. For a comparative examination
of eleven other subnational institutional indicators, see Aseema Sinha, "Divided Leviathan: Comparing
Developmental States in India" (Ph.D. diss., Cornell University, 2000).
35. The library of iNDEXTb contains numerous reports about feasible projects and sectoral informa-
tion.

36. West Bengal did not have a similar agency until 1994, while Tamil Nadu's government established
Guidance, a similar organization, in 1992.
37. Interviews with about forty-five investors in West Bengal and about thirty-five investors in Tamil
Nadu.

38. The classic study is Andrew Shonfield, Modern Capitalism: The Changing Balance of Public and
Private Power (London: Oxford University Press, 1965). Also see Peter Katzenstein, Small States in
World Markets: Industrial Policy in Europe (Ithaca: Cornell University Press, 1985); and Peter Hall,
Governing the Economy (New York: Oxford University Press, 1986). For a recent study that relies on
national frameworks, see Peter Hall and David Soskice, eds., Varieties of Capitalism: The Institutional
Foundations of Comparative Advantage (Oxford: Oxford University Press, 2001).
39. See Judith Tendler, Good Government in the Tropics (Baltimore: The Johns Hopkins University
Press, 1998); Richard Snyder, Politics after Neoliberalism: Re-Regulation in Mexico (Cambridge:
Cambridge University Press, 2001); and Dali L. Yang, Beyond Beying: Liberalization and the Regions in
China (London: Routledge, 1997).
40. One exception is Daniel Treisman, After the Deluge: Regional Crisis and Political Consolidation
in Russia (Ann Arbor: University of Michigan Press, 2001).
41. Victor Nee and David Stark with Mark Selden, eds., Remaking the Economic Institutions of
Socialism: China and Eastern Europe (Stanford: Stanford University Press, 1989).
42. See Jean Oi, Rural China Takes Off: The Institutional Foundations of Economic Reform
(Berkeley: University of California Press, 1999); Alfred Montero, "Delegative Dilemmas and Horizontal
Logics: Subnational Industrial Policy in Spain and Brazil," Studies in Comparative International
Development, 36 (Fall 2001), 58-89; and Snyder, Politics after Neoliberalism.
43. See Shankkar Aiyar and A. K. Menon, "Ransom State," India Today, May 27, 2002, pp. 35-39.
44. Similar methodological modifications are implicitly suggested by Atul Kohli, The State and
Poverty in India (Cambridge: Cambridge University Press, 1987); Subrata Mitra, Power, Protest, and
Participation: Local Elites and the Politics of Development in India (London: Routledge, 1992); Locke;
and Herrigel. A few rare discussions of the cross-national bias in comparative politics are Juan Linz and
Armando de Miguel, "Within-Nation Differences and Comparisons: The Eight Spains," in Richard L.
Merritt and Stein Rokkan, eds., Comparing Nations: The Use of Quantitative Data in Cross-National
Research (New Haven: Yale University Press, 1966); George M. Fridrickson, The Comparative
Imagination: On the History of Racism, Nationalism, and Social Movements (Berkeley: University of
California Press, 1997); and Richard Snyder, "Scaling Down: The Subnational Comparative Method,"
Studies in Comparative and International Development, 36 (Spring 2001), 93-111.

476

This content downloaded from

[Link] on Thu, 21 May 2026 00:23:09 UTC
All use subject to [Link]