Classes, like Ethnic Groups, Are Imagined Communities: A Response to Rao Author(s): Ashutosh Varshney Source: Economic and Political Weekly, Vol. 32, No. 28 (Jul. 12-18, 1997), pp. 1737-1741 Published by: Economic and Political Weekly Stable URL: http://www.jstor.org/stable/4405624 . Accessed: 15/04/2011 22:09

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at . http://www.jstor.org/action/showPublisher?publisherCode=epw. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

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 [email protected].

Economic and Political Weekly is collaborating with JSTOR to digitize, preserve and extend access to Economic and Political Weekly.

http://www.jstor.org DISCUSSION

internalrural divisions, but primarily in terms Classes,Like Ethnic Groups, of and religion which, accordingto Rao, contradictsmy sectoralargument. Are ImaginedCommunities (ii) Are caste and religious identitiesmore basicthanclass considerations in rural ? A Response to Rao The caste- and religion-basedconcept of intra-ruraldivisions, argues Rao, is also Ashutosh Varshney wrong. The pull of caste and religious identities,DDC argues, is blockingthe further growthof rural,sectoral power in India.A THE applicationof class analysis to India's individual interests can no longer be lower versus upper caste constructionof countryside is in need of an intellectual maintainedin the social sciences. politics splits rural India because , renewal. By repeatedly giving rise to, or I would, instead,focus on the second line whetherupper or lower,cut acrossthe town participatingin, the caste- andreligion-based of reasoning that has emerged from the and countryside. Large-scale caste movements,rural India has not only continued theories of ethnicity and nation-building. mobilisation, thus, undermines sectoral to floutthe core predictionsand propositions Thoughthe primary purpose of thesetheories movements,and the same, I argue,is true of the standard class analysis, but more is to show that ethnic and national groups of religiousmobilisation. Class, Rao argues, importantlyfor our purposes,the concept of areimagined communities,2 they have serious capturesintra-rural divisions better than caste class has also undergoneserious changes in implications for the concept of class. Any or religion, and should be given primacy. thelast decadeand a half.The key challenges collectivity that is larger than a village, a (iii) How should one theorisethe relation- to the conceptof class, as we used to practise neighbourhoodor a small organisationis an ship between class power and state it the 1960s and 1970s, have come from the imagined community for it does not allow behaviour?"The first three-quartersof the theoriesof collective action on the one hand face-to-face intimacy. Thus, sectors of the book",Rao contends,"holds no surprises" andof ethnicityand nationalism on the other. economy and classes, like nations, are also (p 1743), contributingnothing to what we Theories of collective action argue that imaginedcommunities.3 An individualdoes already know about India's agricultural classes consist of individuals - hence it not "naturally" feel his class; such policy. Rao has in mind partsof the book must, at the very least, be demonstratedwhy consciousness depends on political where I make a distinctionbetween power as articulated in society and power as individuals would choose to act according mobilisation,public policy, orotherpeople's expressedwithin the state institutions,and to class, not individual, interests.This is an behaviour towards that individual. The explainpublic policy andstate behaviour as analytic imperativebecause individual and implication is not that class analysis is an outcome more of the latter,less of the class interests can often clash.' A landless irrelevant.All thatis requiredis thatclasses former. peasant may be hurt by the decision of the notbe seen as axiomaticallycentral to politics (iv) Canmy return index meaningfullycapture landlordto keep wages stagnant,but it does or politicaleconomy but demonstratedto be returnsto farming? Rao believes that my not follow that it is in his interests to join actually so. index for measuringwhether farm returns a union or party mobilising agricultural J Mohan Rao's vigorous attack4on my have gone up or down since 1970-71 is workers. From an individual perspective, book, Democracy, Development and the seriously flawed. class mobilisation can have a serious cost: Countryside: Urban-Rural Struggles in CLASSESAND SECrORS the peasantmay be dismissed altogetherby India' (DDC hereafter)is admittedlybased the lord, and a dismissal may incalculably on a class perspective, but he is somewhat Rao misreads the basic analytic purpose increase his misery which the resources of blithely inattentiveto the developmentsthat of my book. DDC is not about the internal an agriculturalunion are typically unableto have posed new questions for the concept workings of the countryside.It is about the alleviate. Moreover, and this is the second of class. His critiqueis strikinglyreminiscent impactof the countrysideon statebehaviour/ partof the argument,if the mobilisingunion of Ashok Mitra's claims in Termsof Trade economic policy. That being so, my key did succeed in getting wages raised for the and Class Relations.6The latter was a fine questions required"going inside the state", landless class, the peasant would benefit tractfor its times,but the times havechanged. not going inside the countryside.The Latter anywaywhether or not he participatedin the Theneat simplicity ofa class-theoreticworld, is relevant only insofar as it is necessary to agitation.We may have moralcompunctions as we used to know it, has been shaken by understandthe former, not in and of itself about such "free-riding"behaviour, but let the resilience of non-class economic and Why? us not conflate the normative and the political actors.,and by the corresponding The way an analystslices up the empirical empirical.The landlessworker has an interest theoreticaltrajectories in the social sciences. universe depends on what the analyst is in the benefitsof class action,not in its costs, ThoughRao's attackon my book is wide- asking.Forexample, my currentproject deals andclass benefits, if available,cannot easily ranging, one can identify, in the order of with Hindu-Muslimrelations in six Indian be denied to the landless peasant,but costs theoreticalimportance, four basic disputes. towns, threepeaceful and three violent. The of class mobilisation must ie individually key puzzle of the projectis: why are (i) Shouldclasses or sectors be the unit of borne.A collective and able to live peacefully in some actionproblem thus exists. analysis?Rao agreesthat in the 1970s and For this Behaviour according to class interest can, 1980s the urban-ruraldivide was a highly towns but not in others? question, - and does, take place, but it requiressolving chargedand visible issue in Indianpolitics. the Indianstate or nationalpolitics which the collective action problem. But he arguesthat a sectoralconstruction of is dissected in DDC - provides only the Given the constraintsof space, I will not politicsdoes not requirethe sectoras a unit context within which Hindu-Muslimrela- pursuethis line of inquiryfurther. Wherever of analysis, for India's countryside is tions in the various townr(and the selected required, my book deals with puzzles of internally divided between classes. The neighbourhoods within each towh) are collective action in the countryside. I call sectoralmovements, Rao argues,were only examined.The book, as a result,has a mode attentionto these points only to indicatethat apparentlysectoral but, in reality, class-biased of narration and analysis that is highly an assumption about the identity of class and andagainst the have-nots. I do conceptualise town-based.The questionposed madegoing

Ebonomic and Political Weekly July 12, 1997 1737 inside the neighbourhoods and towns for-itself was the earlierway of statingwhat conditions are met.7According to one such necessary. I have just said. A similardistinction can be condition,it shouldbe demonstratedthat the The sectoral abstractionin DDC is not drawnbetween a nation-in-itselfand nation- observed consequences were anticipatedin meant to deny that divisive issues do not for-itself, or a countryside-in-itself and a the original motivations of the actors. exist in rural India. Indeed, as already countryside-for-itself. Motivations of the participantsshould be indicated,a principalargument of the book Just as Indian nationalists attacked the establishedindependently, not imputedfrom is thatdivisive identitiesof caste andreligion British as a targetthat would unite India as consequences. The latter allows too much cannot easily be overwhelmed by an a nationdespite internaldissensions, a large sloppiness in the social sciences. economic emphasis on urban-ruralissues. numberof India'spoliticians, though not all, Secondly, larger farmers may indeed For Bharat (the villages) to exercise more in the 1970s and 1980s looked for issues that benefit more from price and input subsidies pressureon India (the cities) than it already would make the external,urban target more than the smaller farmers,but what turnson has, the villagers must increasinglyfeel that politically salient than the intra-village that?An unequalbenefit does not mean that being ruralis more importantthan being a dissensions.There have also been politicians those who receive less would not support, memberof caste andreligious communities. who soughtto convince the ruralmasses that or profit from, the movement for higher Rao believes that the very recognition of class was the mostappropriate prism through prices and subsidies. For what may be more intra-ruralcontradictions, even if wrongly whichthey should interpret their experiences. important for the smaller farmers is how conceptualisedin termsof caste andreligion, Comparedto the constructionof a ruralcon- much better they are today compared to invalidatesthe sectoral category called the stituency by politicians since the 1970s - yesterday,not how muchbetter off someone, countryside. pushingthe statefor highercrop prices, input else is at the present moment. There is no Do disputes within a nation - between subsidies, and credit relief - class-based compellingevidence to show thatthe smaller various castes, between language groups, politicians have fared badly. Beyond West farmersview their welfare more in termsof betweenreligious groups, or indeedbetween Bengal and partially Kerala (see below), how bigger farmers are doing, as opposed classes - mean that the concept of a nation class-basedpolitics has been quite inadequate to whattheir own conditionswere until some is vacuous?Do contradictionswithin a class in putting together large rural coalitions. time back.If anything,the availableevidence - between, let us say, upper and middle What does the greater ability of rural suggests that every addition, of benefit, peasantry,or workers in the organised and politiciansto generatean agrarian, as opposed however tiny, makes greater sense to the unorganisedsectors - mean thatthe concept to a class, pressure in the polity indicate? small farmerthan to the biggerfarmer (DDC, of class is of no use? By themselves, internal One argumentcan be thatsectoral construc- pp 128-29). dissensions within classes or nations do not tions simply have greater resonance with Thirdly,much of the class-basedargument makethe concept of class or nationirrelevant. ruralpeople's lives thanclass constructions. depends entirely on an economic under- They simply complicate the analysis of That, according to Rao, is not true. There standingof interests.Even if smallerfarmers classes and nations, presenting analytic is, he seems to suggest, a second method of do notderive as muchdirect economic benefit difficultiesthat must be encounteredbut not demonstrating the centrality of class: a as the larger farmers do, the sense of requiringthat the concept be abandoned. sectoral construction may carry greater empowerment that a movement provides In most empirical research about macro politicalpower than a class construction,but makes their life much easier vis-a-vis the tendencies, large abstractionsare unavoid- economically speaking,sectoral politics has bureaucracy.In several places (TamilNadu, able.Thus, the disputebetween Rao and me class-differentiated results, and therefore 1971; Punjab, 1973; Karnataka,1983; Uttar is in effect about whether,in understanding what appearssectoral is, in actuality,class- Pradesh, 1987, 1993), the ruralmovements the urban-ruralstruggles in India, a sectoral based. "A little class analysis would have arose as a protest against the bureaucracy abstractionis more analyticallyhelpful than gone a long distance",says Rao, in showing tampering with power and input supplies. a class abstraction,for both are conceptual that "efforts to push up producersubsidies This benefit is actually more importantfor abstractionsand neither is self-evidentlyreal. not only benefit (the ruralhaves) here and the smaller farmer than for his richer Which abstractionmakes greaterempirical now but, inso far as such subsidies serve to counterpart. Landlords and rich farmers, sense, and why? fiscally detract from public investment, in given their resources and standing, can Politics is one way of answering this the future as well" (p 1744). Contrariwise, individually negotiate with the bureaucrat. question.We can ask:which constructionof "therural majority here and now are hurtby Standing alone, the smaller farmeris quite theirinterests - caste-based,religious, class- higher food prices" (p 1743). helpless;organised as a collectivity,he wields based,sectoral - sways the masses most?For Chapter 5 of DDC ('Organising the more power. A strictly economic calculus reasons already outlined, all of these Countrysidein the 1980s') anticipatesthis of class benefits typically ignores this, even constructs are imagined communities for argument.It accepts thathigher agricultural thoughevidence for such motivationscan be they are not small groups of face-to-face prices and subsidies benefit some classes shown to exist (DDC, pp 131-32). intimacies.That they are imagined, of course, (the large and middle peasants) more than DDC does maintain,however, that while does not make them any less powerful, for others (small and marginal peasants). On one cantalk with considerable certainty about imaginationscan be awfully serious deter- three grounds, however, it argues that this the sectoralinterests of the landedpeasantry, minants of economic and political action. result alone cannot constitute a basis for large and small, and show that both benefit The point simply is thatan artefact- a class, claimingthat the sectoralmovement is driven from sectoral demands, it is hard to be so a caste, a nation,a sector- must be transfor- by a class-bias. I will come to the special confident about the landless agricultural med into a subjective experience. Large- case of landlessagricultural labour separately. labourers. Are they hurt by higher agri- scaletransformations of thiskind are typically The first counter-argument is purely cultural, especially food, prices? brought about by politics. Through state methodological.Rao commits what may be Rao has no doubt that they are, which policy and/orpoliticalmobilisation, political called a functionalistfallacy. He reads the makesneitherconceptual norempirical sense. leadersseek to constructcoalitions of various motivations underlying a movement from The green revolution strategy combined small units in order to create a larger com- an argument about its consequences. By producer price incentives and new techn- munity which, to begin with, exists only in the methodological standardsof the 1980s ology to increase production. Both Rao imagination.That a class-in-itself has little and 1990s, functionalismis inadmissiblein and I agree that a strategy that relied meaningfor politics until it becomes a class- the social sciences unless some specific more on expanding the regional spread of

1738 Economic and Political Weekly July 12, 1997 yield-increasing technology, and less on their separate non-economic identities" literallyconsumed the first;and (iii) by far producerprice incentives, would be better (p 1745). the mostpopular sectoral movement of the for the landless poor for it would increase They indeedmay, but do they?How often? 1980s,led by SharadJoshi in Maharashtra, employmentas well as make food cheaper. Can we step beyond what may happen and hadby 1990 come undergreat pressure as From this, however, it does not follow, as see what has actually happened? Even in communalism tore into the heart of Rao concludes and I do not, that higher Kerala, where class-based politicians have Maharashtrapolitics and Shiv Senabegan agriculturalprices inevitably hurt the poor. considerable electoral success, it can be toerodeJoshi's base. What had merely raised The welfareeffect of higherproducer prices arguedthat if the fit betweenthe ezhavacaste its headin 1990has by now gonequite far. on the landless depends not only on the and the rural poor had not been so good SharadJoshi was a candidatefrom two higher consumer prices of food (which an between the 1930s and 1940s, class seatsin the 1993 assembly elections in increase in producerprices, in the absence mobilisationwould have made little headway. Maharashtraand lost both. In the 1996 of offsetting consumer subsidies, would Class politics was insertedinto the campaign parliamentaryelections, SharadJoshi fought typicallytrigger), but also on theemployment for caste-basedsocial justice.'2To this day, fromNanded, widely viewed as a stronghold createdvia higherproduction and the impact the ezhavas continueto be the principalbase of his peasantorganisation. While the Shiv on wage rates.How exactly these three factors of the CPI(M): People of similar class- Sena-BJPcandidate polled 1,73,366votes, work themselvesout - food prices, employ- positions, if nair, have gone on the whole Joshigot a mere71,407. A communalShiv ment, wages - is an empirical question. with the Congress;if Christian,with Kerala Sena-BJPalliance appears to have fatally Regional variations are likely to exist. Congress; if Muslim, with the Muslim woundedJoshi's organisation and political ConsiderPunjab. In 1965-66, about 31.5 League.'3 platform. per cent of the ruralpopulation was estima- If casteand religion have not been displaced If such evidence does not present a ted to be below the poverty line, consisting in Kerala where class-based mobilisation falsifiabilitytest for-the greater pull of caste mainly of the landless. By 1970-71, this has achieved some success, what can we say andreligion, I amnot sure what does."3 Had proportionhad droppedto 17.5 per cent and aboutthe rest of the country?A class analyst Raobeen more familiar with India's political by 1990-91, still furtherto 3.45 per cent.8 could still arguethat the vibrancyof religious universe,he would have both seen that a FromBihar, however, declines as large have and in Kerala (or elsewhere) falsifiabilitytest exists, and that it hasbeen not been reported. is a faultof the strategiesdeployed by Marxist met.Caste-politics and religious nationalism Evidence from outside India further politicians, not a comment on "the reality". are fightingit out for nationalprominence complicatesthe empiricalpicture. Studies of Such a position, if taken,would not resonate today, pushing class-based,or sectoral, Malaysia show that the first phase of the well with what we know from the field, and politics to the margins,except in small greenrevolution (1967-74) led to remarkable will only invitethe charge that we arearrogant quartersof India. risesin welfareall around:"it is a rarepeasant ivory-towertheorists, who believe we know POWERAND THE today who does not eat rice twice a day", the realities better than the best organisers CLASS STATE reportedJames Scott from Sedakain 1974.9 do. In a disarmingly candid statement, Like his understandingof class, Rao's However, the introduction of combine- E M S Namboodiripadhasrecently admitted commentson the relationship between classes harvesters,by far the single largest labour- that the inability of the decades-long class and state policy also have a touch of displacing machine in agriculture,created mobilisation in Kerala to overwhelm the theoreticalobsolescence. Even according to agriculturalunemployment by the late 1970s. religious divisions of the state may be rather Marxists,Marxist theory of the state was Withina decade, peasantfortunes declined. more rooted in historical realities than reductionistin the 1960s: state behaviour No such complexity enters Rao's argu- Marxists had expected.'4 As people who used to be directly deduced from the ments aboutthe landless. He is deductively spend less time in the field organising, let correlationof class forcesin the economy sure that the poor are hurt, but the claim is us not second-guess the most towering, or society.Therefore, reacting in the 1970s, entirely premised on the argument that tireless, lifelong mobiliser of the working NicosPoulantzas developed a theoryof the higherfood prices hurt the landless poor, classes in independent India on what the relativeautonomy of the state.'6In liberal withouttaking into accountthe othereffects. "realities"are. circles,too, the state theory was reductionist A similar reasoning is applied to all net Rao further argues that my hypothesis untilthe 1960s,except that state behaviour buyersof foodgrain,including small peasants. aboutthe greaterpower -of caste or religious was"read off" from interest groups, not from Deductive reasoning based on one effect factors vis-a-vis the sectoralconstruction of classes,and interests,unlike classes, were alone is not enough to ascertainwhich way economic interestsis "unfalsifiable".This is consideredcross-cutting, which thereby gave the results would go for the poor,"' making a surprisingclaim. The argumentis not only permanentpolitical advantage to no particular unambiguousjudgments about their wel- testable,but DDC also presentsthe following interestgroups. By the 1970s,liberal state fare nearly impossible. As it turns out, the evidence in support (pp 186-88): (i) the theory,too, had movedin the directionof case studies of who participates in price sectoral, peasantagitation in Punjabled by "bringingthe state back in", giving up movements do reflect this ambiguity, BKU gathered increasing strength till the reducingstate behaviour from societal groups showing,that some landlessworkers support mid-1980s, forcingmany concessions on the and pressures.'7 the movement while others oppose it." government,but the religious mobilisation Rao seemsunperturbed, unconvinced, or triggered by the Sikh insurgency after the unmindful. He sees nothing analytical or CLASSES AND IDENTITYGROUPS mid-1980s literally wiped out the sectoral new in the first three quartersof DDC, in Let me now turn to the relative salience movement; (ii) when, in the campaign for which I track down the history of struggles of class, caste and religion in ruralIndia. My the 1991 general elections, three different over agricultural policy within the state argumentabout caste andreligious identities constructionsof India'sbasic conflicts were institutions and provide an explanationfor weakening a potentially greater rural presented to the electorate as competing the outcomes thatemerged, refusingto infer (sectoral) pressure in the polity is wrong- choices for the futuremap of Indianpolitics state policy and behaviourfrom conflicts in headed, thinks Rao, because "people who - the urban-ruraldivide led by Devi Lal, the society.As faras I know,no politicaleconomy have identical economic interests vis-a-vis upperversus lower caste constructionled by book on India,with the exceptionof Francine other classes in society may unite to press V P Singh, and a Hinduversus Muslim Frankel'swork in the 1970s,'8has gone inside theirjoint economic claims notwithstanding constructionby L K Advani- thelatter two the state institutionsto analyse the struggles

Economic and Political Weekly July 12, 1997 1739 conducted over economic policy. Frankel in contrast,constitute the centrepieceof the of P, C and the appropriateprice detlator, and I, however, apply different conceptual micro-theory of prices. By now, scholars the magnitudeand "even the sign of changes and methodological prisms to the same have soughtto bridgethe macro-and micro- in R" (Formula 2) may "differ from those empiricalsite. A second widely readtreatise theories of agriculturalprices by importing of Rt" (Formula 1). by PranabBardhan identifies three dominant the notion of border prices in agricultural The first objection does not make sense. classes - the landlords,the bourgeoisie and policy debates.20Between the early 1950s C in Formula 2 is not equal to what the the bureaucrats- but Bardhanderives state and late 1970s, however, the macro and government of India calls total costs (or policy from the presumedinterests of these microperspectives on agriculturalprices were Cost C), which include operationalcosts as classes andthe equilibriathey generate.'9 He strikingly at odds. Because they constitute well as land rents. It only includes opera- does not investigate battles over economic the argumentative arsenal of economic tional costs (or what the government calls policy within the state institutions. bureaucracies, international and national, Cost A),22as all graphs on farm returnsin Rao's theory of public policy and state economic theories matterin public policy. my book repeatedlymake clear (pp 162-64). behaviouris straightforward.India's state, Inferringpublic policy from class power in I did not use Cost C on purpose.If we include he says, has been in the grip of the "rural civil society is simply wrong. land rents as costs, we run into a serious haves" on agriculturalpolicy: conceptual problem. Are land rents price- THE RETURN INDEX Theinfluence of largelandowners on pricing determiningor price-determined?Or, to use policymust be seento be of a pricewith their Rao also critiquesmy farmreturns index, DavidRicardo' s famousformulation, is price influenceon variousother policies...Long which I deploy to measure whether farm of corn high because the price of corn land beforethe rise of theheadline-grabbing lobby returnsover time have gone up or down. is high, or is the price of corn land high orientedtoward price policy, the ruralhaves Using the agriculture-industryterms of trade because the price of corn is high? The latter hadgiven a good account(though not public data to derive farm returns,DDC argues, is was true,argued Ricardo, because land supply display)of theirability to defend their interests misleading because the terms of trade only was relatively fixed. Land rents, thus, can againstthose of the ruralhave-nots. Both reportrelative prices, not costs. Depending be seen as "pure rents" on an inelastically landreform and tenancy reform were scuttled on costs, farm returnscan go up even when suppliedfactorof production. Forthis reason, by their money, muscle and access to agriculturalprices decline relative to industry. C in my formuladoes not include rentsover governmentmachinery (p 1744). This is especially truein periodsof technical which, then, the governmentwould provide Even as a description,let alone theory,the change, when increases in yield reduceunit a margin in its administeredsupport price claim that the small class of "ruralhaves" costs. (P). That I "cannotbe faulted for supposing is behindthe failureof land reformsas well There are two ways of factoringin costs. that what is good for the pure rentieris also as the relative success of price movements The ideal formula would be: Rt (farm good for the poorestrural labourer" (p 1744) is highly dubious.What does the term"rural returns)=(P-C)Y (where P is output price, is, thus, an ideologically overdetermined haves"mean? If it means both landlordsand C unit cost, and Y yield per acre). Let me conclusion. We can conduct arguments the so-called middle peasantry,then it can call it Formula1. However, to measurereal without such ideological one-upmanship. be shown that the former benefited from as opposed to nominalreturns, this formula Class analysts, I suspect, are not the only blocking land reforms, but the latter were requiresa pricedeflator. As the fierce debate people who care for the poor! either hardly touched by land reforms (if over the appropriateprice deflator for the The second objection, in principle, has theirfarmsizes werebelow the size-ceilings), agricultural sector indicates, there is no greatersubstance, though it is not clear what or they were the beneficiariesof reforms(if uniquelyacceptable deflator available in the its overall impact is. It is true that the they were substantialtenants of the absentee economic literature.2'It is not a problem directionalityof returnsbased on a deflated landlords).Thus, even if their interests are scholarsof politicaleconomy can easily solve. (P-C) and of those based on (P/C) can differ. similar on agriculturalprices today, they Until economists settle the debate, we must But that will be true if and only if the werenot on landreforms. Moreover, whether willy-nilly go for the second best. proportionatechange in P is greater than theinterests of landlordsand middle peasants A second-best solution was available proportionatechange in C, but the absolute are similar or conflictual, we still have to because my argument did not require change in C is greater than the absolute show how these interests are played out measurement of exact returns, only change in P. My sense of the cost and price inside the state institutions if we wish to directionalityof returns.Whetherfarm returns datafor 1970-90 is thatthis is a most unlikely understandstate policy. Should one theorise were going up or down as a result of the condition. If Rao wants to show that it is about state policy without asking what rising agrarianpressure in the polity was all not only conceptually possible but also happens inside the state institutions? I had to find out. I, therefore, constructed empirically true, he must investigate the Once we go directly to the institutionsof a returnindex, which I wrote as R=(P/C)Y publicly available data for 20 years. Is Rao policy-making,we notice, as DDC shows, (where R is return index). Let me call it the significance of ideas and economic Formula2, which eliminatedthe problemof theories.The power of large landownersin finding an appropriateprice deflator, for P, civil society notwithstanding,their biggest a nominal magnitude,was being divided by adversaryinside the state in the 1950s was C, another nominal magnitude, and could, the development theory of the time. The if multiplied by yields (Y), indicate Teachers/Students:Rs. 300 latter,in its searchforquick industrialisation, directionality.On the basis of this index, I Hardbound:Rs. 800 looked primarily at the macro-economic found that farm returns were neither Paperback:Rs. 500 consequencesof agriculturalprices. Not until uniformly increasing or decreasing. They did the landowners penetrate national went up or down, dependingon the crop and Write to: parliament and the policy-making the state. Institute for Studies in was institutions, lasting political legitimacy Rao has two objections to my proposal. Industrial Development given to the micro-economic view of First,my indexis a mere"proximate estimate NarendraNiketan, I.P. Estate agricultural prices. The macro view has of land rents obtainedafter materialand all New Delhi - 110 002 traditionally been hostile to the idea of labourinputs have been deducted"(p 1744). email: [email protected] producerprice incentives; such incentives, Second,dependi ng on therelative movements

1740 Economic and Political Weekly July 12, 1997 preparedto do that? All of us perhaps will 'Farm Prices and Class Bias in India', EPW, the Special Expert ,Committeeon Cost of learn. Review of Agriculture,September 1979. Production Estimates, Governmentof India, In conclusion, let me summarisethe main 22 For details, Ministryof Agriculture,Report of Delhi, 1980. thrustof my response.In political economy, if not in economic theory, simple-minded deductions can be highly vacuous, and if ImportTariffs as StrategicPolicy Signals combined with ideological certitudes,they canalso misdirectand undulypolarisedebate. Murali Patibandla Deductions and polemics are no substitute for careful empirical work. Our conceptual IN reply to,my note in this journal (1995) excellentempirical example ofthis possibility work must mesh with the real world. on importprotection and exports, Marjit and is given in a recent book by Jacobsson and Sarkar(1995) havecome out with a technical Alam (1994). By comparingthe Indianand Notes point of optimal pricing rule of monopoly SouthKorean hydraulic excavators industry, instead of addressingmy main contentions. they point out that "...while the Indian 1 JonElster, Making Sense of Marx,Cambridge In this note I would like to restatemy main government liberalised its UniversityPress, New York, 1985; and Adam licensing policy Przeworski,Capitalism andSocialDemocracy, contentionsand take this opportunityto put in the 1980s to allow for a greaterlevel of CambridgeUniversity Press, New York, 1986. forwardan interestingpoint of using import domestic competition, the Korean policy 2 Benedict Anderson, ImnaginedConmmunities, tariffs as strategic policy signals towards meant a restrictionin the field to only two Verso, London, 1983. disciplining domestic producers. producers.Second, while in the case of India, 3 Yael Tamir, 'The Enigma of Nationalisin', Vast theoretical and empirical literature the protection from imports seemed WorldPolitics, April 1995. on industrial organisation and trade in indefinite, the Korean government clearly 4 J MohanRao, 'BeyondUrban Bias', Econiomic imperfectcompetition framework shows that set a limit on the protection."In otherwords, and Political Weekly(EPW hereafter),July 6, 1996. analysis of exports should not be based on restriction of entry into Korean industry 5 CambridgeUniversity Press, New York, 1995; piecemeal approaches because any single allowed firms to realise economies of scale and FoundationBooks, Delhi, 1996. trade or industrialpolicy has strong trade- advantagein exportsand the threatof import 6 Frank Cass, London, 1977. offs. High levels of importprotection could competition made them make systematic 7 See the debate between Jon Elster, 'Further work against export promotionif it causes technological efforts to remain highly Thoughts on Marxism, Functionalism and long-rundomestic marketpower and makes competitive.In other words, the South Korean GameTheory,' and G A Cohen, 'Marxismand domestic producershighly X-inefficient in policy was able to combine trade and Functional Explanation' in John Roemer, production.In the case of Japanand South industrialpolicies quite effectively. It was Analytical Marxism, Cambridge University Press, Cambridge, 1986. Korea,import protection was combinedwith able to use import protection levels as an 8 H S Shergill and GurmailSingh, 'Poverty in a set of industrialpolicies underan effective effective strategic signalling towards dis- Rural Punjab', EPW, June 14, 1995. institutionalmechanism of 'carrotand stick' ciplining domestic producers.The policy in 9 WeaponsoJ the Week,Yale University Press, policies. Indialed to the opposite results:the delicen- New Haven, 1985, p 66. My second contention is a technical one sing led to excess entry and to suboptimal 10 George Blyn, 'The Green Revolution and has interestingimplications in terms of scales and the policy signal of indefinite Revisited', Economic Development and usingimport tariffs as strategicpolicy signals. importprotection to inefficiency [Patibandla Cultural Chlange, 1983. July Accordingto Marjitand Sarkar,at a dmestic 1995a]. If export promotion is the 11 Jim Bentall and Stuart Corbridge, "Urban- policy RuralRelatilons, Demand Politics and the 'New price of Pw(]+t), where Pw is world price objective, it has to be pursuedby an effective Agrarianism' in North-west India: The and t is importtariff, whole of the domestic combinationof a set of trade and industrial BharatiyaKisan Union", Transactionsof the market will be catered to by the domestic policies by minimising the negative trade- Instituteof BritishGeographers, 21 (1), 1996. monopolist.At anygiven time,if thedomestic offs, ratherthan by piecemeal approaches. 12 Dilip Menon, C'aste,Nationialism and Co,n- producerfollows the price of Pw(l+t), the munism: Malabar, 1900-1948. Cainbridge domestic market structure is no more a References University Press, Cambridge, 1994. monopoly as imports also face the same 13 T J Nossiter, Communismin Kerala, Univer- Jacobsson, S and Ghayur Alam (1994): sity of California Press, Berkeley, 1982, price. At price Pw(1+t), both domestic Liberalisationand IndustrialDevelopment in pp 345-75. producer and imports share the domestic theThird World, Sage Publications,New Delhi. 14 EMS Namboodiripad.'Presidential Address', market. Marit, S and A Sarkar (1995): 'Protection and International Congress of Kerala Studies, One way the domestic monopolistcan be Exports', Economic and Political Weekly, August27-29, 1994, AddressesandAbstracts, shown to cater to the whole of domestic September 16. Vol 1, AKG Centrefor Researchand Studies, market at price Pw(l+t) is by introducing Patibandla, M (1995): 'Import Protection and Trivandrum,1994, p 2. timeelementinto the model.This is illustrated Exports',EconomicandPolitical Weekly,July 22. 15 For conceptualarguments along this line, Jon as follows. Let,us take two time periods I Elster,'Three Challenges to Class', Analytical -( 1995a): 'South Koreaand India:A Tale of Two Marxism. and 2. At period l, the domestic policy gives Development Experiences', International 16 Political Power and Social Classes, New Left a signal to the domestic monopolist that at JournalofDevelopmentBanking, Vol 13,July. Books, London, 1975. a futuretime period 2. importtariffs will be 17 Theda Skocpol et al, Bringing the State Back reduced.This promptsthe domestic producer In, CambridgeUniversity Press, New York, to commit a capacityor sales in the domestic Economic and Political Weekly 1985. marketas anentry deterrence strategy towards 18 India's Political Economy, Princeton blockingthe entryof importsat period2 even Availablefrom University Press, Princeton, 1978. at the price of Pw( 1+t). In this case, exports 19 ThePoliticalEconom)y(sofDevelopmnentinhIdia, Sainath Book Centre will if Blackwell, Oxford, 1984. decline the cost curve remains G- II, MahendraChambers 20 e g, Anne Krueger, The Political Economyof unchanged between the two periods. The V T MagazineMarket Agricultur-alPricing Policy, Vol 5, The Johns interesting point here is that cost curves Dr D N Road Hopkins University Press, Baltimore, 1992. might shift down as a responseto the policy Mumbai - 400 001 21 Thecontroversy was touchedoffby D S Tyagi, signal of reduction in import tariffs. An

Economic and Political Weekly July 12, 1997 1741