Relatively developed and progressive south Indian states–Tamil Nadu and Kerala—went to the polls in May to elect new governments. North Indian states known to be relatively backward, such as Uttar Pradesh and Uttarakhand, elect their legislative assembly in 2017. Are there lessons the latter can learn from the southern states to climb the ladder of development more rapidly than has been the case thus far?

There are several journalistic observations about the divide between the northern and southern Indian states, as Paul and Sridhar highlighted The Paradox of India’s North-South Divide: Lessons from the states and regions (Sage, 2015).  In the 1950s and 1960s, there was an exodus of south Indians to “north Indian cities” such as Delhi, Kolkata and Mumbai, in search of jobs.

South Indians (or “Madrasis”) were required to go to the “north” to improve their lot. But the tide of migration has in the 2000s shifted in favour of south Indian cities such as Bengaluru, Chennai and Hyderabad.

Rising above the journalistic accounts and observations, we can be more formal in this analysis and examine trends in economic performance indicators such as per capita income for the northern and southern Indian states. Paul and Sridhar (2015) present the historical trends in per capita Net State Domestic Prodcut (NSDP) in the south1 and north Indian states (see chart overleaf), which showed a clear divergence between the two groups of states beginning from the 1990s.

The widening of the gap between the two groups of states began after the mid-1980s and became more pronounced after 1992-93. Based on this analysis we conclude that judged by per capita income the southern states were always ahead of the northern states by a modest margin, but that they moved far ahead of their northern counterparts by 2005. The two regions differed only by 39 per cent in terms of per capita income in 1960-61 while the gap had widened to 101 per cent by 2004-05.

info 1.jpg
In the case of individual southern and northern states, Paul and Sridhar report that in 1960-61, Tamil Nadu (TN) had a per capita income of ₹5,053 against Uttar Pradesh’s ₹3,338, with TN being ahead by 51 per cent. The gap narrowed to 39 per cent in the early 1980s. However, by 2005-06, the gap between the two states in terms of per capita income had widened to 128 per cent. The economic gap between the two groups has thus widened significantly in recent years, and shows that the surge of the south is indeed a recent phenomenon.

Is the surge of the south a paradox? If yes, what explains this phenomenon? There are many aspects of this “paradox”, i.e., the dramatic turnaround of the south. The magnitude is noted by Datt and Ravallion (1998) who reported nearly 70 per cent rural poverty for Tamil Nadu in 1960 compared with only about 48 per cent for Uttar Pradesh, which is now considered backward. This suggests that economic deprivation and inequality were much higher in TN earlier on, but that it made a surge in terms of reducing them rapidly at some later point.

What explains the economic divide between the two groups of states? While Paul and Sridhar (2015) attempt many explanations for this paradoxical divide, including human capabilities, skills, awareness, urbanisation, infrastructure, developmental spending and its efficiency, and governance, in this article, I concentrate on selected aspects of the economic divide between the two groups of states—including access to basic public services, access to private assets, efficiency of spending, and governance, in a limited sense.

Why do some states do better than others at providing services? There is one instance that is relevant. In Bihar’s Tirhut Range, a mystery illness strikes mainly children every year in summer, extending to the middle of May. It kills dozens of small children on average and this has been happening for more than two decades, but as yet there is neither a cure nor even a health department protocol to be followed. The encephalitis epidemic of Gorakhpur district, UP, in 2005, is another instance.

Contrast this with Kerala’s handling of Chikungunya. It, too, has no cure but the health services have worked out a regime that optimises the treatment of a patient. It was discovered only in the 1990s.

So why do we observe such a difference with the way in which basic services are provided? This is only one instance, but there are many others in various sectors that could be highlighted—school enrolment, reducing female infanticide, the mid-day meal scheme, some of which Paul and Sridhar (2015) extensively discuss.

I shall attempt to answer the following questions: Does a higher growth rate of per capita income translate into better basic services for the public and better quality of living, as reflected in their assets? I examine the following basic services for which the data are available from the Census: drinking water, electricity, sanitation (latrine), bathing and lighting facilities. I also examine access to assets, which reflects on quality of living.

A look at the data on basic public services in the two reigions shows the difference clearly. In the summary of data for 2001 and 2011 the graphic overleaf shows that on negative traits such as percentage of households that did not have a latrine facility on their premises, households with waste water outlet not connected to any drain, and the proportion of households with kerosene as their main source of lighting, the southern states registered a decline in varying degrees, which is to be expected with increasing incomes and better governance.

info 2.jpg

On traits which are negative, namely lack of access to sanitation in one’s house, lack of access to a sewerage system (open or closed), or the use of polluting fuels such as kerosene for lighting—the northern states registered an increase during 2001-11, which implies an unfavourable change during the period. The percentage of households using kerosene for lighting was small (in relative terms) in the southern states, being one-fourth as of 2011, whereas this was more than 50 per cent in the northern states in the same period.

In the case of basic facilities such as wastewater connected to a drain, bathrooms in the house, electric lighting, and running water at home, northern states recorded an improvement in the decade. In fact, there was a more than 100 per cent increase in the percentage of households with bathing facilities in the northern states, but even this increase (Census 2011), covered only 65 per cent of households. More than 75 per cent of households in the southern states had a bathing facility in-house (at least within the premises).

When we examine electricity for lighting only 79 per cent of households in the northern states had it, while more than 92 per cent of southern households used electricity as the main source of lighting.

It is possible that the efficiency with which resources were utilised in the southern states failed the northern states. Further, the actual delivery of public services is determined by several factors—high literacy rate accompanied by a high degree of civic activism, household income and preferences, the economies of scale determined by density, the in/efficiency with which such public services are delivered, which possibly determine better delivery of public services. Given that the southern states are more literate, have a greater degree of civic activism, with higher per capita income, and more recently, better governance, it must have been the case that the southern states were better equipped to deliver basic amenities than their northern counterparts.

In order to answer the question whether a higher growth rate translates into a better quality of life, we attempt to summarise the possession of private goods (assets) by households in the northern and southern states.

The percentage of households with assets such as a radio/transistor, television and landline phone is unambiguously higher in the southern states than in the northern states (see graph on following page). Especially given that the landline phone is relatively more time consuming to install (Sridhar and Sridhar (2007)), we expect only permanent and higher income households to have access to this.

info 3.jpg

In both groups of states, the proportion of households with radio/transistors registered a decrease, which is consistent with increased use of mobile and smart phones and television sets, which showed substantial increases in both groups of states, while higher in the southern states, as we expect, with their higher per capita income.

Thus we find that a higher growth rate in the southern states has also translated into a better quality of life for them, both in terms of basic amenities and possession of assets, whereas a lower growth rate has prevented the residents of northern states from enjoying a better quality of living.

What explains better delivery of public services in the southern states? A common explanation offered for the surge of the south Indian states and the consequent better delivery of public services, and access to private assets, is governance. If we were to compare the governance of states, it is easy to imagine that, where chief ministers’ tenures are longer, investors, who are the key agents of economic growth, would expect stability in policies, and the regime overall would be stable and sustainable. So any state where the tenure of CMs is long is said to be governed better. Even here, the paradox shows up in a comparison between northern and southern states.

Paul and Sridhar (2015) noted that the tenure of chief ministers of the south Indian states was in fact lower than that of the northern states in the 1960s (Figure 1). So it cannot be claimed that south Indian states were better governed from the beginning. The paradox shows then that historically, even in something as basic as governance, things have not been favourable to south Indian states. However, governance in terms of the CMs’ tenure changed in favour of south Indian states in the decade of the 2000s, along with its associated benefits.

Another hypothesis is that the south Indian states spend more on development. Per capita development expenditure could be important as it results in the creation of productive assets. Paul and Sridhar (2015) examined trends in public investment and per capita developmental expenditure in the southern region versus that in the northern states, respectively summarised in Figures 2-3.

While there is a lot of variability in per capita developmental spending across states in the northern region when compared with those in the southern region, it is important to remember, that spending without attention to outcomes, may yield less value for the spending envisaged. Attention to the efficiency with which resources are utilized, and the consequent developmental outcomes are necessary.

Given there are no significant differences in per capita development spending across the two groups of states, in order to examine the efficiency of resource use, Paul and Sridhar examined expenditure on sectors such as roads and education which are inputs, and that on respective outcomes such as the change in road length and learning, across one southern state, Tamil Nadu, and one northern state, Uttar Pradesh.

Recognising that outcomes manifest themselves only with a lag after the initial expenditure/investment, in the case of roads2, Paul and Sridhar used the 1980-85 period for examining expenditure and with a five-year lag, 1985-90 for the outcome, i.e., road length. Tamil Nadu spent a total of ₹92,483 during 1980-85 for creating every additional kilometre of road during 1985-90, whereas UP spent a lot more, 3.5 times more, i.e., ₹3,28,788 over 1980-85 to create an additional kilometre of road during 1985-90.

Therefore, in the case of roads, given their relative record of spending, Tamil Nadu fared better than UP in efficiency. For a given budget, it was able to build more kilometres of road than UP. This could be due to fewer officials who monitored the works presumably more effectively, fewer labourers who worked the full hours expected of them, due to which the southern state was presumably relatively more efficient than its northern counterpart.

Paul and Sridhar took another example from primary education to demonstrate the efficiency of resource use in the case of the same two states. They reported about a couple of surveys by the Public Report on Basic Education in India (PROBE) team in the Hindi-speaking states (in 1996 and 2006), which showed that despite the fact that schooling infrastructure had expanded rapidly in these states, classroom activity levels had not improved during the decade.

The surveys found that there was an impressive increase in the number of primary schools between 1996 and 2006, with one out of every four government schools being set up during the decade. The number of schools in UP with at least two pucca rooms rose from 26 per cent in 1996 to 84 per cent in 2006. In 1996, free uniforms were provided only in 10 per cent and textbooks in less than half of schools. This increased, respectively, to more than half and nearly 99 per cent of schools in 2006 in the Hindi-speaking states, all of which were found to be impressive in terms of inputs.

If we relate the inputs discussed above to outcomes, as Paul and Sridhar did, in rural north India, in 1996, about half the time, there was no teaching in primary schools. Despite all the increases in resources and inputs during 1996-2006, a second survey by PROBE in 2006 found that nothing had changed with respect to educational outcomes–half the government schools still had no teaching activity when the investigators arrived, in the Hindi speaking states.

info 4.jpg

What is the evidence from south India? The Annual Status of Education Reports (ASER), conducted by Pratham, in 2006 found that the percentage of children out of school in TN was 4.9 in the age group 7-16, 3.6 in the age group 11-14 and 15.8 in the age group 15-16 (both boys and girls), compared with 8.9, 8.9 and 22.6 respectively for UP that year. This is despite the fact that TN’s proportion of spending on elementary education during the period 1994-95/2009-10 was a meagre 1.67 per cent of the total spending on education. Again, as in the case of roads, it is possible that teaching actually happened in TN, whereas it likely did not happen in UP.

There is reason to believe that the educational outcomes observed in TN and the southern states are better generally because of the mid-day meal programme, as reported by Paul and Sridhar (2015), based on extensive discussions with bureaucrats of that state who were in power during what we believe must have been the transformational period.

In TN, the mid-day meal scheme was part of the overall investment in human capital, in health-cum-education, during the 1980s. A big part of this story was the mid-day meal scheme initiated by the then chief minister M. G. Ramachandran’s regime. Sridhar and Singh (2003) also reiterated the benefits of the mid-day meal scheme, which has been further supported by eminent economists such as Jean Dreze.

As reported by Paul and Sridhar (2015), the focus in the mid-day meal programme was not only just feeding, but also on nutritional improvement. There was a regular practice for doctors visiting government pre-schools to monitor the nutrition intake amongst children to ensure that there was the right combination of carbohydrates, proteins and vegetables in TN. The local population was also involved in the programme whenever it was possible, and their suggestions were included to make the programme locally relevant. This model was subsequently replicated across the country. Such initiatives in south Indian states such as TN must have certainly contributed to the positive educational outcomes there.

Summarising, the lessons for the northern states which are set to go to polls in 2017, from the southern states, are to improve their investment in human capital —both health and schooling, more importantly to monitor all developmental outcomes with better governance and efficiency, elect a leader/political party who are likely to remain longer in power, and lend stability to key economic actors.


1. The south refers to the states of Tamil Nadu (TN), Karnataka (Kar), Andhra Pradesh (AP), Telangana (Tel) and Kerala (Ker). The north consists of Uttar Pradesh (UP) (along with Uttarakhand which was created in 2000), Bihar (Bih) (along with Jharkhand created in 2000), Madhya Pradesh (MP, along with Chhattisgarh carved in 2000) and Rajasthan (Raj) (also originally known as BIMARU states), consistent with the categories in Paul and Sridhar (2015). I do not comment regarding western Indian states such as Maharashtra/Gujarat vis-avis the southern states.

2. It should be noted that the mix of roads (for e.g., rural roads, national and state highways), the use of material, the regional cost of the material used (e.g., cement, asphalt versus others), land terrain, and other factors could make a difference to the cost of roads per km, possibly accounting for a part of the cost differential noted above.

Developmental expenditure by the new states of Uttarakhand, Chhattisgarh and Jharkhand have been added to that of their parent states—respectively Uttar Pradesh, Madhya Pradesh, and Bihar, to ensure comparison in the pre-2000 and post-2000 periods.

This article draws extensively upon the book co-authored with Padmashri late Dr Samuel Paul, on the Paradox of India’s North-South Divide, published by Sage in 2015.



Datt, Gaurav and Martin Ravallion (1998) “Why Have Some Indian States Done Better than Others at Reducing Rural Poverty?” Economica 1998, 65, 17-38.

Paul, Samuel and Kala S Sridhar. The paradox of India’s north south divide: Lessons from the states and regions, New Delhi: Sage, 2015.

Sridhar, Kala Seetharam and Shailendra Singh. “Knowledge-hungry yes, food-hungry no”, The Hindu Business Line, October 3, 2003.

Sridhar, Kala Seetharam and V.Sridhar. “Telecommunications Infrastructure and Economic Growth: Evidence from Developing Countries,” Applied Econometrics and International Development, 7 (2) (2007): 37-56.