Statistical numbers tell us many things, but it is also important to read them in context and not jump to the wrong conclusions. From the 3.5 per cent 'Hindu rate of growth' of the mid-20th century to 8 per cent early in the 21st might seem like an enormous leap.

However, in an important sense the change is less dramatic than the numbers suggest--because the structure of the economy has changed in the intervening half a century and this is heightening the growth effect, whereas each sector of the economy is not growing much faster than it was half a century ago.

To make the point clear: in 1950, agriculture accounted for 55 per cent of GDP; now its share is down to barely 20 per cent. This means that slow agricultural growth has much less of an impact on GDP now than half a century ago--which is one reason why droughts are not as devastating as they used to be.

At the same time, services have nearly doubled their share of GDP, climbing from about 28 per cent then to approximately 54 per cent now. And since this is the fastest-growing sector, its impact on overall GDP too is magnified.

If we were to apply today's sectoral growth rates for agriculture, industry and services to the economy and the sector shares of the 1950s, GDP growth would have been a little over 5 per cent, not 8 per cent.

This is not to downplay the importance of having reached 8 per cent GDP growth in the first half of this year, after averaging 7.6 per cent in the last two years. Instead, it is intended to show how structural change in the economy underpins the new numbers, and to emphasise that there is no real sector-wise acceleration.

The point is important, because agriculture per se has not accelerated its rate of growth; if anything, it has decelerated to 2 per cent annual growth from as much as 4 per cent in the 1980s because of slow productivity change and under-investment.

Similarly, industry used to grow at 7 per cent even in the Nehru years, and has done so through the last decade as well. The country's industrial momentum therefore has not undergone any real change, except in the last couple of years.

But since the share of industry has grown, GDP gets an additional boost. The same broad statement goes for services, though not entirely so in this case.

What would add credit to India's economic performance (creditable enough as it already is) would be to make each sector accelerate its pace of change. That is what would signal a true transformation of the economy.

If agriculture were to double its pace to the 4 per cent growth rate of the 1980s, for instance, the impact on GDP would not be huge, resulting in only 0.4 per cent faster overall economic growth. The same impact on GDP would be achieved if services accelerated from the 10.1 per cent growth rate achieved in the last quarter to 11 per cent growth.

But the impact in the countryside, on both employment and poverty levels, would be much more dramatic in the first case because agriculture still absorbs more than a third of India's labour force, and accounts for the bulk of India's really poor people. Also, the spread effects of a doubling of agricultural growth would be infinitely greater.

The flip side of this, of course, is that the structural change in the economy will continue, as the faster-growing sectors account for ever increasing shares of GDP.

So, even if nothing else changes, overall GDP growth will continue to accelerate and could reach 9 per cent within a decade, even without any other change in the economy.

The other cheerful aspect of the last three years' economic record is that it is significantly better than what had been forecast in Goldman Sachs' BRICs report of 2003, where all that was postulated was that Indian GDP growth in the second half of this decade would be 6.1 per cent! Clearly, reality is set to trump the forecast.