I have for long been a believer in the economic determinants of voting in India, i.e. people vote with their minds, and vote for performance. But being an empirical junkie, there seem to be so many exceptions to this rule that I have started to reluctantly believe in the theorem that performance is definitely not one of the determinants.

Electoral success has many fathers, and we may find it difficult to ascertain the contribution of each. But we can definitely rule out the contribution of performance.

This theorem applies only to incomparable India; in western democracies, it is likely that economic performance matters a lot.

It is a mistaken belief in the economic model that has led various experts to identify reasons for Lalu's success in Bihar, with one enterprising TV reporter even going to the extent of saying that Lalu had won in the past because Biharis had chosen dignity over development. And now he had lost by a landslide because after 15 years of misrule, dignity had exhausted its surplus!

The "performance equals votes" model can be examined both on an all-India basis, and at the more micro level, the state. After 40 years of Nehru-Indira Gandhi economics, economic reforms were introduced in India, reforms which have helped to double the per capita growth rate, and in the last three years, to triple the performance of the pre-1990s.

Prior to 1991, the per capita growth rate averaged around 2 per cent per annum; with a present GDP growth rate of 7.5 per cent per annum, and a population growth rate of only 1.5 per cent, it is simple maths, which yields a per capita growth rate of 6 per cent.

But what did the Indian voter do in response to her betterment? She threw out the Congress in 1996, and threw the party out convincingly. In 1991, the Congress won 232 seats; in 1996, 140 seats. Interestingly, there have been three national elections since then, and the Congress has reverted to the charismatic dynastic leadership, and yet it has not been able to exceed the 1996 tally by more than 5 seats. Proof that neither dynasty nor performance matters!

But the Indian voter was not done, yet. The Vajpayee-led National Democratic Alliance, after successfully steering the Indian economy to the next stage, e.g. acceleration to the 7 per cent plus GDP growth, lost the 2004 elections.

These two events (its own continuous losses and the NDA's loss as well) were enough for the Congress party leadership to believe that what the Indian voter really wants is an even more populist government than that provided by Indira Gandhi in the 1970s; hence, the justifiable claim by the Sonia Gandhi-led United Progressive Alliance to be the most populist in the world this side of Hugo Chavez of Venezuela.

Lalu Prasad ruled Bihar for 15 years; Messrs CPM have ruled West Bengal for 28 years. If economic performance mattered for elections, then surely these two states must have done well; not only well, but better than the average state.

For a large set of economic performance indicators, for the 1980s and 1990s, this is just not the case. In terms of per capita growth, West Bengal is average, and Bihar ranks 14th out of the 17 major states.

In terms of change in mean years of education (a priority for the Communists and populists), WB is 14th and Bihar a much better 10th! For change in gender equality, Bihar does again better - its rank is 6th compared to a low 13 for the pro-people, pro-gender CPM's West Bengal.

Lest one believe this is due to any lack of "expenditure" (populists in the UPA, please take note), growth (between 1983 and 1999) in real state expenditure per student was the highest in West Bengal, 160 per cent, compared to 107 per cent for Bihar, and only 33 per cent for the rest of India!

Land reform is often talked about as the need of the hour, for both vote-getting and economic goals (higher growth). The CPM's claim to fame is that in a country in desperate need of land reform, they are the only ones who provided it.

For the period 1980 to 2000, per capita state domestic product in agriculture, nationwide, grew at only 1.1 per cent per annum, West Bengal grew at 3.2 per cent and Bihar at -1.5 per cent. Surely, proof of why CPM keeps on going like the Energizer bunny. However, with a considerably worse land distribution, and no land reform, Rajasthan PCSDP grew at 2.9 per cent per annum.

It cannot be ruled out that voters actually vote for performance, but the intervening, and more important, determinants help swing the balance. What are some of these variables? Probably, the most important is political thuggery, e.g. voter intimidation, beating up opposition candidates, stuffing electoral booths, etc.