Discourses on globalization

back to issue

A SPECTRE is haunting the world’s governments – the spectre of globalization. Some argue that predatory market forces make it impossible for benevolent governments to shield their populations from the beasts of prey that lurk beyond their borders. Others counter that benign market forces actually prevent predatory governments from fleecing their citizens. Although the two sides see different villains, they draw one common conclusion: omnipotent markets mean impotent politicians. Indeed, this formula has become one of the cliches of our age. But is it true that governments have become weaker and less relevant than ever before? And does globalization, by definition, have to be the nemesis of national government?

Globalization is a journey. But it is a journey toward an unreachable destination – ‘the globalized world’. A globalized economy could be defined as one in which neither distance nor national borders impede economic transactions. This would be a world where the costs of transport and communication are zero and the barriers created by differing national jurisdictions have vanished. Needless to say, we do not live in anything even close to such a world. And since many of the things we transport (including ourselves) are physical, we never will.

This globalizing journey is not a new one. Over the past five centuries, technological change has progressively reduced the barriers to international integration. Transatlantic communication, for example, has evolved from sail power to steam, the telegraph, the telephone, commercial aircraft, and now the internet. Yet states have become neither weaker nor less important during this odyssey. On the contrary, in the countries with the most advanced and internationally integrated economies, the ability of governments to tax and redistribute incomes, regulate the economy and monitor the activity of their citizens has increased beyond all recognition. This has been especially true over the past century.

The question that remains, however, is whether today’s form of globalization is likely to have a different impact from that of the past. Indeed, it may well, for numerous factors distinguish today’s globalizing journey from past ones and could produce a different outcome. These distinctions include more rapid communications, market liberalization, and global integration of the production of goods and services. Yet, contrary to one common assumption, the modern form of globalization will not spell the end of the modern nation state.

Today’s growing integration of the world economy is not unprecedented, at least when judged by the flow of goods, capital and people. Similar trends occurred in the late nineteenth and early twentieth centuries.

First, the proportion of world production that is traded on global markets is not that much higher today than it was in the years leading upto World War I. Commerce was comparably significant in 1910, when ratios of trade (merchandise exports plus imports) to GDP hit record highs in several of the advanced economies. Global commerce then collapsed during the Great Depression and World War II, but since then world trade has grown more rapidly than output.

Mainstream economic thought promises that globalization will lead to a widespread improvement in average incomes. Firms will reap increased economies of scale in a larger market, and incomes will converge as poor countries grow more rapidly than rich ones. In this ‘win-win’ perspective, the importance of nation states fades as the ‘global village’ grows and market integration and prosperity take hold.

But the evidence paints a different picture. Average incomes have indeed been growing, but so has the income gap between rich and poor countries. Both trends have been evident for more than 200 years, but improved global communications have led to an increased awareness among the poor of income inequalities and heightened the pressure to emigrate to richer countries. In response, the industrialized nations have erected higher barriers against immigration, making the world economy seem more like a gated community than a global village. And although international markets for goods and capital have opened up since World War II and multilateral organizations now articulate rules and monitor the world economy, economic inequality among countries continues to increase. Some two billion people earn less than $2 per day.

At first glance, there are two causes of this divergence between economic theory and reality. First, the rich countries insist on barriers to immigration and agricultural imports. Second, most poor nations have been unable to attract much foreign capital due to their own government failings. These two issues are fundamentally linked: by forcing poor people to remain in badly governed states, immigration barriers deny those most in need the opportunity to ‘move up’ by ‘moving out’. In turn, that immobility eliminates a potential source of pressure on ineffective governments, thus facilitating their survival.

Since the rich countries are unlikely to lower their agricultural and immigration barriers significantly, they must recognize that politics is a key cause of economic inequality. And since most developing countries receive little foreign investment, the wealthy nations must also acknowledge that the ‘Washington Consensus’, which assumes that free markets will bring about economic convergence, is mistaken. If they at least admit these realities, they will abandon the notion that their own particular strategies are the best for all countries. In turn, they should allow poorer countries considerable freedom to tailor development strategies to their own circumstances. In this more pragmatic view, the role of the state becomes pivotal.

Why have economists and policy-makers not come to these conclusions sooner? Since the barriers erected by rich countries are seen as vital to political stability, leaders of those countries find it convenient to overlook them and focus instead on the part of the global economy that has been liberalized. The rich countries’ political power in multilateral organizations makes it difficult for developing nations to challenge this self-serving worldview. And standard academic solutions may do as much harm as good, given their focus on economic stability and growth rather than on the institutions that underpin markets.

Prasenjit Maiti


Tamil Nadu – the numbers game

THE recently concluded elections in Tamil Nadu not only gave a sweeping mandate to the AIADMK led front, but also sent all the psephological projections for a six. An outright majority for Jayalalitha left all political pundits dumbfounded; they had predicted that either of the Dravidian parties which rode to power would have to depend on sub-regional parties and caste-based outfits to secure a majority in the state assembly. Besides swings in the mood of the electorate, the arithmetic of alliances has come under close scrutiny after the May 2001 polls. The debate on the arithmetic of alliances has also kick-started the debate on the role of smaller parties and the transferability of their votes.

A number of surveys have tried to obtain an estimate of the support bases of various parties. These bases minimally need to arithmetically add up for the success of the alliance. However, contradictory interests of support groups and local factors, such as lack of cohesion between cadres of allying political parties have often whittled the gains expected at the time of conception of the alliance. Often, neither of the parties in the alliance benefits, i.e. none of them is able to augment their seat tally, sometimes with a loss in the percentage of votes polled.

The BSP-Congress alliance in Uttar Pradesh for the 1993 polls is a case in point. Neither the BSP nor the Congress managed to increase their final tally significantly, despite the alliance. In the light of these experiences the elections in Tamil Nadu provide us with an opportunity to study the politics of alliances as both the DMK and the AIADMK had forged together alliances of political forces and ideologies which did not exactly resonate with each other. Jayalalitha’s success establishes her as an astute political manager for three reasons: one, her ability to weld together all the non-NDA political formations in the state with a proven mass base into one front; two, her ability to wrest the most winnable seats, often at the cost of annoying many sitting MLAs; three, apparently, a complete transfer of votes amongst the alliance partners.

While the AIADMK concluded its seat sharing pact (decision on the number of seats to be contested, not the actual allocation) by the 17th of March, Karunanidhi was bargaining with smaller parties until the second week of April. Experiences of previous elections have shown that seat sharing deals that are concluded faster leave enough time for the alliance to gain acceptability in the lower echelons of the member parties as also help build up effective cadre coordination. Hastily stitched alliances have often been counterproductive. Being almost a month ahead in stitching her alliance Jayalalitha effectively won the first round of the battle.

Jayalalitha’s major victory has been her success at exploiting the psychological weakness and political helplessness of her allies. The PMK quit the NDA after being assured the chief minister’s post in Pondicherry. Going back to the NDA was a sealed option as far as the immediate elections were concerned. The Pondy carrot ensured that the PMK did not bargain too hard in Tamil Nadu. Of the 27 seats allotted to the PMK, nine were DMK bastions and nine were seats where it did not have a very strong base to boast of. The TMC dreaded the idea of a third front after it proved to be a non-starter in the 1999 Lok Sabha polls. The INC and the Communists have hardly any significant base to win even a couple of seats as part of a third front. It was this helplessness which Jayalalitha exploited to the hilt and extracted the best seats for her party in the deal.

Out of TMC’s 39 sitting MLAs only 24 could be renominated from their respective seats. The CPI was forced to give up five of the eight seats it held then. The Congress for the first time did not contest any seat in the Tirunelveli district. Seats in DMK strongholds such as Chennai city were left by and large for the allies. The AIADMK contested only five seats in Chennai city (including one TMK candidate who fought on the AIADMK symbol). Thus, after having won the first round of the battle the AIADMK supremo ensured that she got the best seats, often risking the repercussion of unfair seat sharing deals and allocations from workers of alliance partners. Her initial offer of five seats to the Congress was an indication of her intention to drive a hard bargain with that party. A statistical analysis of the seats won by member parties of the DMK and the AIADMK fronts throws up enough quantitative evidence to establish how overloaded the latter alliance was in favour of Jayalalitha’s party and that the seat distribution in the DMK led front was probably more even.

A total of 2,80,44,840 valid votes (including postal votes) were cast in Tamil Nadu for the polls held on 10 May 2001. Of these the AIADMK led front polled 1,39,95,381 votes (49.9% of all the valid votes). The DMK led front polled 1,08,42,470 votes (38.67% of all the valid votes). The break up of the two fronts is as follows.




Vote share (%)



















* includes the TMK candidate who contested on the ADMK symbol and that of one independent


DMK Front


Vote Share (%)





PT (Pudhiya Tamizhagam)


MTD (Makkal Tamil Desam)


NJP (New Justice Party)


MGRK (MGR Kazhagam)


Thondar Congress


CNMK (Congunadu Makkal Katchi)






TB (Tamizhar Bhoomi)




* Here the vote share of the DMK includes that of the DPI,TMC (Democratic), Muslim Eikya Jamaat, Muthariyar Sangam, Farmers and Toilers Party, PMK (Dheeran) and AIMMK whose candidates contested on DMK symbols.


The average vote polled by each of the parties in the seats that they contested compared with the statewide average vote polled by their respective fronts is an indication of the relative winnability of the seats allocated to a particular party as compared to the other members of the front, i.e. if the state wide average vote share of the AIADMK front is 49.9% but the average vote share of the AIADMK in the seats that it contested is 52% then we can argue that the AIADMK managed for itself a greater share of the seats where the alliance was placed comfortably. However, the strike or success rate (number of seats won divided by the number of seats contested) may not be a statistically accurate criterion as victories or losses could be a result of lower or higher IOUs respectively varying from constituency to constituency.

We now take a look at the average vote share of each party of the AIADMK front in the seats that it contested. The plus sign next to the average vote share of the party in the seats that it contested indicates that its share in its own seats is higher than the state average while the negative sign indicates that the party’s average vote share in the seats that it contested is less than the statewide average of the front of which it is a member.

Average vote share of AIADMK front members in the seats that they contested


Avg. vote share of the party in the seats that it contested


52.00 (+)


48.54 (-)


48.2 (-)


47.48 (-)


46.82 (-)


45.9 (-)




The above table clearly illustrates that the AIADMK is the only party with a vote share higher than the statewide average of the front. All the other member parties are below the state average of the front. This is one indication of the winnability of the seats offered to these parties, clearly demonstrating the lopsided distribution of seats by Jayalalitha. A strong counter argument, however, emerges from the NDTV-Frontline-CSDS post-poll survey. While many of the member parties of the AIADMK front were forced to sacrifice their choicest seats, in terms of popularity Jayalalitha’s AIADMK was way ahead of the other member parties such as the PMK and the Congress. In that sense these parties have been beneficiaries of the AIADMK’s munificence and their sacrifices could be justified. On the other hand the DMK front represents a more even distribution with some members of the alliance being above and the rest below the state average of the DMK front.

Average vote share of DMK front members in the seats that they contested


Average vote share of the party in the seats that it contested (%)


40.8 (+)


40.79 (+)


39.98 (+)


39.02 (+)


38.74 (+)

Thondar Congress

38.6 (-)


37.16 (-)


33.75 (-)

New Justice Party

33.50 (-)




32.4 (-)

State Average



In the DMK led front it can be noted clearly that five members of the alliance are above the state average of alliance in terms of vote share, while six members are below. Parties like the MTD, BJP and DMK contested seats that were probably more winnable than the rest. The rest of the parties polled below the state average of the front. Thus, it may not be wrong to conclude that the seat distribution of the DMK led front did more justice to its member parties than the AIADMK front.

Another measure could be the percentage of margin of victory of the candidates of the various parties, though this may also be affected by the IOU (Index of Opposition Unity). The average percentage of margin of victory of these parties is a rough indicator of the toughness of the battle they had at hand. The two tables below illustrate how misleading these numbers can be. The following table indicates average percentage of the margin of victory of various members of the ADMK front.


Average percentage of margin of victory of the ADMK front members


Average percentage of the margin of victory














The unexpectedly high average percentage of the margin of victory of the Congress indicates the flaw in this method. This high margin results from two assembly constituencies where the figure was as high as 25% and 30% respectively. If these two constituencies are excluded then the average percentage of margin of victory for the Congress candidates turns out to be a mere 11.18%.

A similar table for the DMK front is shown below.

Average percentage of margin of victory of DMK front members


Average percentage of margin of victory








Here too, the average margin of victory is higher for the MADMK because of a relatively high margin of victory in one constituency. Therefore, the percentage of margin of victory is not an accurate indicator of the quality of seats allocated to a party at the time of seat allocation, and may be used only as a starting point for further analysis.

Jayalalitha’s shrewd bargain paid off and resulted in an absolute majority for the AIADMK. What would the scenario have looked like had it not been for this tough seat allocation deal? Alternatively, what would the assembly have looked like had Jayalalitha doled out the seats evenhandedly to her allies, that is a proportionate share of weak seats for all the members of the alliance including the AIADMK. The table below gives an estimate of the number of seats, which could have been won by each party in case of an evenhanded deal.



Existing deal

Even handed deal




















The number of seats, which the AIADMK would have won in an evenhanded seat allocation is 118, the magic number required for a simple majority. A single seat less and she would have been vulnerable to pressure from her allies. This explains her strong arm tactics in dealing with most of her allies. Had it not been for the large number of allies she brought into her fold, a high IOU against the DMK would not have been possible. This would have ensured fewer seats for her alliance in the new assembly and greater dependence on the allies for the ADMK. By raising the IOU against the DMK, Jayalalitha ensured that her front secured a near landslide victory. Then, within the limited share of seats available for allocation she carved out a share for herself, which included the most winnable seats. This ultimately translated into a simple majority for the AIADMK.

Her overwhelming victory has made analysts overlook an important fact. Despite a landslide for her alliance, Jayalalitha has barely managed to secure a simple majority. In any future election an anti-incumbency sentiment could easily take her below the halfway mark. Moreover the best bargain in terms of seat allocation, like the 2001 elections, may not always be possible. This could probably be her last cakewalk.

In addition to deconstructing the arithmetic that made Jayalalitha’s alliance succeed, it is necessary to understand the increasingly important role of smaller outfits in Tamil Nadu which will decide the destiny of these alliances. It is here that the debate on transferability of votes is located. Animosities on the ground often prevent effective social alliances, though they may produce winning combinations on paper. The recently concluded election in Tamil Nadu saw the emergence of a number of caste based outfits in the state. Most of these were allies of the DMK. Prominent among them were the PT (Pudhiya Tamizhagam, DPI (Dalit Panthers of India), MTD (Makkal Tamil Desam) and the NJP (New Justice Party).

The PT (Pudhiya Tamizhagam) of K. Krishnaswamy first contested elections independently as a political party in 1998. In the southern districts it enjoys a strong following among the Pallars. The party contested 14 Lok Sabha seats and polled an average of five per cent of the valid votes in these seats. The PT did extremely well in the southern constituencies of Ramanathapuram, Sivakasi, Tirunelveli and Tenkasi polling 9%, 13%,14% and 19% of the votes respectively. Of these the party contested Ramanathapuram, Tirunelveli and Tenkasi in 1999 in alliance with the TMC. While the gains due to the alliance were modest in Ramanathapuram and Tirunelveli (4% to 6%), the PT made handsome gains in Tenkasi polling 34% of the vote. While this data is insufficient to conclude as to what percent of votes of other parties like the TMC are transferable to parties like PT, it does establish that in alliance with mainstream parties the PT could be a deciding force in assembly segments under these parliamentary constituencies.

In the 2001 elections the PT lost all 10 seats it contested including Ottapidaram and Valparai, the two seats from where Krishnaswamy contested.It is still a matter of debate whether effective social coalitions can be built between the social constituencies of other parties and that of the PT. However, of the four PT seats, which are a part of Ramanathpuram, Tirunelveli and Tenkasi, three were lost because the MDMK carried away a significant portion of the anti-ADMK vote. Had it not been for the MDMK, the PT probably would have won three seats. Since the MDMK hurt the chances of the DMK front in certain districts of southern Tamil Nadu, it would be unfair to blame the PT alone for its poor show. The loss to the PT is a part of the larger damage inflicted upon the DMK front by the MDMK and should not necessarily be inferred as PT’s inability to attract voters of allying parties. The table below lists the seats that the party lost due to the MDMK’s presence.


Assembly constituency

Margin of defeat for PT (%)

Vote share of MDMK (%)











The BJP benefited greatly due to the absence of MDMK candidates. Had the same been the case with the PT, it would have won three out of the 10 seats it contested (strike rate 30%), given the fact that the BJP and the DMK have been able to win just 19% and 17% respectively of the seats that they contested. However, the PT’s performance outside Ramanathapuram, Tirunelveli and Tenkasi has been very poor. Here the average margin of defeat of a PT candidate is above 20%. This probably indicates the geographical limits of PT’s mobilisation.

The DPI (Dalit Panthers of India) contested eight seats on the DMK symbol. The party has a strong following among the Parayars in northern Tamil Nadu. DPI leader Thirumavalvan narrowly won from the Manglur constituency. In Cuddalore and Chidambaram, where the DPI has a substantial following, the DMK front candidates romped home in five out of the 12 assembly segments at stake. In a very limited way though, the DPI did manage to help the DMK resulting in a gain of about half a dozen seats.

The MTD (Makkal Tamil Desam) was formed by S. Kannapan, a minister accused of corruption in the earlier Jayalalitha regime. The party has been projected as being sympathetic to the cause of the Yadavas. Though the party did not win a single seat its performance has nevertheless been far more enviable than the other allies in the DMK front. The party contested five seats on its own symbol and polled 40.8% of the votes, which is higher than the state wide average of the DMK front of 38.67%. Moreover the average margin of defeat of MTD candidates was only 5.8%, commendable given the fact that the average margin of victory throughout the state is more than twice this. The party lost three seats by margins that were less than five per cent of the votes. This again includes one seat where the damage was done by the MDMK. A marginal swing in favour of the DMK alliance would have seen the MTD open its account in the assembly. In that sense the MTD cannot be discounted from future calculations.

The NJP, projected as a party of the Mudaliars, indeed cut a sorry figure. Four out of five seats allocated to the party in the DMK alliance were in northern Tamil Nadu, a DMK bastion. Inspite of this, the party did not win a single seat, registering an average margin of defeat of 17%. In Bhavani the party candidate was defeated by a margin close to 30% of the vote. The party’s average vote share in the seats that it contested was a mere 33.5%, even less than that of the PT, which performed quite badly in a few seats.

Thus, most of the caste based outfits have a significant nuisance value and could play a decisive role if the two Dravidian fronts are evenly balanced in future elections. It would be a mistake to write them off, going by the poor show in terms of the number of seats they won.

Though Jayalalitha has managed to obtain a simple majority for her party and ensured a government on her own, it looks highly improbable that it will be possible to repeat such a feat. While both the Dravidian parties have around 31% of the votes , it is the fragmentation of the non-DMK and non-AIADMK vote which has given birth to a plethora of smaller parties. Unlike the past where the Congress enjoyed a solid block of 18% to 20% of the vote, there are now a host of parties competing for about 35% of the votes not committed to either of the Dravidian parties. Thus every Dravidian party now needs to evaluate the marginal value of the vote of the party, which would be incorporated into the alliance.

By marginal value of vote one means the extra number of seats that can be won for the front by incorporating a party with an estimated vote share into the alliance. Beyond a certain point, adding more parties (marginal votes) into the front might not add more seats to the alliance. This concept is similar to that of marginal utility in economics. Unlike the past where the incorporation of the Congress into the alliance ensured a one-sided mandate, the battle now is for every extra fraction of the vote (represented by the smaller parties) that can drive you a few seats ahead of your rival. Therefore, the current task of a Dravidian party is not to enlist the support of just one party as in the past, but to add up as many small parties, who summed together would form a huge block of the vote, ensuring a decisive lead over the rival front.

Jayalalitha established an early lead in this race . Karunanidhi too added many parties to his front, but did not get the best pick. But the sheer number of such parties in the future will ensure an even race, as the system will always tend towards equilibrium. A political polarisation like Kerala with two fronts could soon emerge; either of the fronts being led by a Dravidian party. Then would dawn the era of true coalition politics in Tamil Nadu as landslides would be a thing of the past. The results of the 1998 and 1999 elections are harbingers of things to come. The recently concluded elections are only an aberration.

Jai Mrug