How deep is the divide between the economies of urban India and the rural "Bharat", where most of our population lives? Bits and pieces of data coming out of various agencies suggest that not only is the chasm deep, it has grown substantially over the past few years.

There are vast disparities in growth rates, and the link between the two economies is at best tenuous. What it augurs for the sustainability of growth and indeed for the future of Indian politics is a question that policy makers, politicians and investors in India need to answer urgently.

Let me start with the evidence. According to the 2001 census, 62 per cent of India's population is rural. They would, by definition, depend directly or indirectly on agriculture for their livelihood. If I look at the Mid-Term Appraisal of the Tenth Five-Year Plan, which the Planning Commission has recently released, I find the following.

The trend rate of growth in agriculture declined from the 3.2 per cent that it had notched up in the period between 1980-81 and 1995-96 to just 1.9 per cent in the subsequent period. Interestingly enough, the deceleration in growth is not confined to just a few crops or rural population like the small and marginal farmers.

It is a general phenomenon spread across crops and "represents a broad based decline in productivity growth". It is not surprising that per capita income for the rural population has actually declined over this period.

Let me do some quick arithmetic here. We keep talking of a minimum aggregate GDP growth of 6-7 per cent in India. If agriculture constitutes roughly 25 per cent and is growing on average at 1.9 per cent, the rest of the economy, comprising industry and services, has to grow at least 7.3 per cent.

Thus we have a dual economic structure in which a 1.9 per cent growth economy supports more than 60 per cent of the population and a 7.3 per cent growth economy supports the remaining 40 per cent. I will refer to the two segments as the 1.9 cent and 7.3 per cent economies.

This model of growth would be sustainable if the faster-growing economy systematically absorbed population from the slower one. As the pressure of population on the latter declines, productivity and real wages should rise, leading to some sort of equalisation across the sectors.

The Lewis model (the brainchild of West Indian Nobel prize-winner Arthur Lewis), which had a profound influence on policy makers for a couple of decades, talked of a growth process based on this transfer of labour surplus to the labour-scarce manufacturing sector, in which the dualism would finally atrophy. The Chinese growth model also follows a variant of this.

What are the links between the 1.9 per cent economy and the 7.3 per cent economy, of rural and urban India? Is the latter doing its bit to take the pressure of population off the low-growth economy? Not quite. The MTA figures show that organised sector employment both in the public and private sectors declined in the period 2001-03. I suspect this trend would have persisted earlier since most manufacturing companies' balance sheets show a decline in the wage-sales ratio beginning way back in 1998-99.

The service sector, as we know it, has always been a bit of a non-starter when it comes to employment. Let me take the case of the much-touted BPO sector. It currently employs about 400,000 people.

Even if the BPO sector were to grow at a scorching 40 per cent rate over the next few years, it would employ only about 1.2 million people by 2010. Ditto for IT and telecom. In any case, given the kind of skills that are needed in these sectors, it is quite unlikely that they have much need for migrants from the rural areas.

The only link between the high- and low-growth economies is then through the large unorganised sector that operates on the fringes of the high-growth economy, such as construction and household work.

Anyone who has seen how migrant workers live in construction sites should be well aware of the fact that these aren't exactly the hottest jobs going. Not only are they poorly paid, they are also highly seasonal and uncertain. In that sense, the bulk of urban informal sector jobs are not much different from rural employment.

Given this scenario, the policy imperatives are somewhat obvious and I will resist the temptation to pontificate on them. I would recommend the Planning Commission MTA as an excellent reference for those interested in the specifics. I am a little more interested for what it means for manufacturers and service providers setting up shop in India.

The first and obvious conclusion is that the effective size of a particular market may turn out to be far smaller than assumptions based on broader macro numbers might suggest. Thus, companies might find themselves hitting saturation points much sooner than expected. This is particularly critical for those whose forays are taking them beyond the 7.3 per cent growth economy and into the 1.9 per cent growth markets.

Based on this hypothesis, let me hazard a guess on the future of cellular telephony. The first flush of penetration of cellphones was entirely focused on the urban population. The growth in penetration was relatively easy not just because tariffs came off sharply but also because the entire target population came from the 7.3 per cent economy.

The next round of growth will have to come from the rural or the 1.9 per cent. My guess is that after an initial spurt (driven by the small numbers of rural affluent), penetration rates will slow down suddenly as average income and affordability levels nosedive.

From a broader strategic perspective, it might make sense for an investor looking at India to confine himself entirely to the higher-growth segment. However, while growth in income and affordability might be strong, competition might be equally fierce.

The reason is obvious: since that's pretty much where all the action is, everyone and is uncle is trying to get into that space. So pricing power is unlikely to be high -- growth in these markets will come at the cost of profit margins. The experience of white goods producers, mortgage lenders, car manufacturers -- indeed any sector that has seen high growth rates over the last few years -- should bear me out on this.

But as management guru C K Prahalad keeps pointing out, there's a fortune at the bottom of the pyramid as well. I guess he is talking of the 1.9 per cent economy, where despite falling per capita incomes and virtual stagnation, the sheer number of people does offer opportunity for business.

But the rules of the game are radically different in these markets. As you can imagine, affordability, transaction sizes and price-sensitivity are likely to be very different from the high-growth economy. The bottom line is that the dual structure throws up clear distinct strategic choices for those trying to sell in these markets. Those who are getting a little carried away by the India story need to be aware of how different these options are.

The author is chief economist, ABN Amro.  The views here are personal.