The GDP estimates for the second quarter of 2003-04 (July - September) released on the last day of last year provided lots of New Year's eve cheer.

GDP grew at 8.4 per cent over the corresponding quarter of last year. For the historically minded, this is the second time in the brief history of quarterly GDP data (the series was first put out for 1996-97) that the growth rate has exceeded 8 per cent.

The only other time was in the second quarter of 1998-99, 8.2 per cent, when the community, social and personal services component grew at over 20 per cent, presumably as a result of payments of Pay Commission- related arrears to government employees.

Apart from these two quarters, in no other quarter since the series began has growth even crossed 7 per cent. Obviously, the boom years of the mid-90s are not captured, but we have to work with the data we have.

Thus, the 8 .4 per cent mark is not only the highest ever, but also, more importantly, not the outcome of any one single, dominant factor. Looking at the sectoral patterns that have contributed to this record, a number of interesting stories emerge.

The most obvious is the revival of the agricultural sector, in response to the quality of the monsoons in most parts of the country. Last year, this sector had declined by 3.5 per cent during the second quarter; this year's positive growth of 7.4 per cent represents a virtual return to the long-term trend of about 3 per cent in this sector.

Looking at the immediate future, this suggests that the 8.4 per cent record will be short-lived, because in the third quarter of last year, agriculture declined by 7.6 per cent.

Since the monsoon performed right through its duration, a significant boost for agriculture in the third quarter, with a growth rate of around 10 per cent, is on the cards. Assuming that the other sectors persist with their rates, growth in the third quarter of this year will turn out to be closer to 9 per cent.

But, at the same time, looking ahead to the next year, the very quick recovery of agriculture back to its trend this year implies that it will not contribute very much to growth next year.

A normal monsoon in 2004 will support the trend rate of 3 per cent, a deficient monsoon even less. Unless the other sectors accelerate beyond their current performance levels, we will still be subject to the tyranny of the base effect.

In any case, agriculture is about 20 per cent of GDP and its gyrations will matter less and less to GDP growth. The more interesting stories underlying this quarter's growth are in the other sectors.

Among the components of industry, the positive story is in manufacturing, which accounts for about 18 per cent of GDP. This quarter, with 7.3 per cent, marks the fifth successive quarter with growth above 6 per cent. It also provides the first signs of acceleration beyond this pace.

For those of us who track the monthly Index of Industrial Production, an interesting transformation has occurred in the relationship between the IIP and the GDP in manufacturing. During the boom years of the 1990s, the former outgrew the latter by a considerable margin. For some time after that, the two moved closely together.

In the last few quarters, GDP growth has been accelerating beyond production growth. This suggests the dominance of productivity over volumes as source of industrial growth. The question is whether the persistence of 6 per cent-plus over the last year and more will redress this balance.

The indications are that it will, through a revival of private investment; in which case we can expect the performance of the manufacturing sector to improve in the coming year.

One negative indicator on the industrial scene is the sluggishness in the electricity sector. It (along with water and gas) accounts for only about 2.5 per cent of GDP, but its growth rate of below 3 per cent during the quarter, and that too declining, is clearly at odds with the acceleration in manufacturing. Is this because of increasing dependence on captive power, which is inadequately captured by the data system? Or is it something else?

A well-known critique of China's growth numbers revolves around the inconsistency between energy consumption and the volume of output in the economy.

Does the discrepancy between the two growth rates in our economy cast doubt on the robustness of the manufacturing numbers? In any case, if nothing else, it underlines the potential for the power sector to become a binding constraint on a sustained industrial recovery.

Services continue to show a remarkable buoyancy. Accounting for 52 per cent of GDP, they grew at almost 10 per cent during the quarter. A sector with this combination of high share and high growth rate will obviously provide an enormous boost to overall growth.

Looking at the components of this sector, the virtuous share-growth relationship clearly rests on trade, hotels, transport and communication.

This component contributes about half of service sector GDP and grew at almost 12 per cent during the quarter. It was also the fastest growing component by far during the previous quarter. It now accounts for about a quarter of GDP, almost equal to the entire contribution of industry.

Trade and transport are clearly linked to the total volume of goods produced and moved, in both agriculture and industry. In the communications segment, perhaps this spurt is linked to the expansion of the market, which resulted from the introduction of CDMA services in the early part of calendar 2003. Community, social and personal services also beat the overall GDP growth rate.

The ability of the service sector to remain completely immune from the cycles that have plagued both agriculture and services during the last few years is something that needs to be understood far better than it is now.

A reasonably high-frequency set of numbers for this sector would allow us to decompose the various demand and supply factors that have contributed to this outcome. Are any of them replicable in the other sectors, which might lead to at least a moderation of the volatility we have seen recently?