Friday, May 23, 2014

How big is Indian Middle Class? From Laptops to Motorbikes: 10-20%




In my previous post, Malik posed an interesting inquiry- by Global standards, how big is Indian/South Asian middle class.  How large is the size of what we may call the ‘haves’ in this region?

I’ll focus here on India – and pray pundits may do the number crunching for the rest of South Asia or perhaps correct or refine the estimate I propose for India. The object here is not to arrive at an exact figure but a rough estimate whose upper and lower bounds are fairly certain. 


What Census Data tells us

The latest, most reliable and most comprehensive data on Household possessions for India  is the Census 2011 Data . While a middle class household must possess a decent house with all the ‘basic’ global middle class amenities- a modern toilet, kitchen, electricity supply, water supply and so on; Census data does not provide us such holistic picture ( no data on number of households having ALL these basic amenities) but certain key indicators (which one may term as ‘elite’ and ‘Aam Aadmi’ indicators) can be used to arrive at a rough estimate of Middle Class household share in populace. 


Upper Middle class & elite Households 

 Census data on 3 of the ‘elite’ indicators:

o    Households owning Car/Jeep/Van:  4.7%

o    Households owning Computer/Laptop: 9.4%

o    Households having Mosaic/Floor tiles:  10.8%

Assuming any household which owns any of the above 3 items must fall under middle class, one can reasonably claim that middle class size should be atleast 10% of all households (afterall how many poor households can spare money for Laptop/Computer or Mosaic/Floor tiles? Very few I believe).


Mainly middle class members:  Significant chunk, if not Majority of these should fall under middle class)

Census data on some of the “Aam Aadmi” (Common middle class) indicators:

 o    Households owning Scooter/motor cycle/Moped: 21%

o    Households  owning LPG/PNG for cooking: 29%

o    Households owning Concrete roof: 29%

o    Households owning Cemented floor: 32.2%

Again assuming any household who doesn’t own any one of the above items doesn’t deserve middle class status, one can reasonably claim that middle class households should not be more than around 20% of all households (afterall how many middle class households can’t afford even a scooter/motor cycle/moped? Leaving out environmentally conscious, Very few I guess).

Putting the two bounds together one can reasonably claim that going by Census 2011 Data:

 Middle Class households account for 10-20% of India’s all households.


I admit this is a very simplistic analysis (cheaper recycled stuff; elite/middle class separation; traditional cow dung users and many other details overlooked) devoid of intense academic rigor; and can be refined further but very broadly, It does give us a reasonable rough estimate which also sits well with estimates coming from  multiple other sources, for example:





PS: Is it an overestimated, underestimate or needs to be refined further (closer to 10%, or to 20% or somewhere in b/w)? Comments are most welcome.

4 comments:

  1. Thanks for creating this thread, Naveen. I have researched this subject to some extent before but the difficult part is that there isn't much of a consensus among sources as they tend to indicate very different numbers. For example, in 2012 Christian Meyer and Nancy Birdsall came up with the following technical note in which they estimate that by global standards the Indian Middle Class is about 70 million people(using a cutoff of a 10 dollar a day income).
    http://www.cgdev.org/doc/2013_MiddleClassIndia_TechnicalNote_CGDNote.pdf

    This is the lowest estimate I have ever seen bandied about and there were one or two people who seemed to have disagreed with their estimate. If the cutoff is reduced to at least an 8 dollar a day income then the size of the middle class rises to 91 million people. Of course, they also excluded anyone with a greater than 50 dollar a day income so if we add this designated 'upper class' to the middle class numbers the number of people who are comparable to first world status is larger(unfortunately they did not include the estimate of the number of people who earned more than $50 a day).

    Many older sources estimating middle class sizes in different countries would use a cutoff of only $2 a day(so barely escaping the dismal dollar a day poverty statistic) in order to be included in the middle class and thus you would get numbers such as 400-500 million people constituting India' middle class. While this is one extreme, I wonder if the 70 million estimate is the other extreme(too conservative)? Here are a few more articles by Birdsall regarding India's middle class and growth in the size of the population earning between $4-10 a day(Birdsall predicts the size of this class to be the one to include the largest number of people in India within the coming decades).
    http://www.cgdev.org/blog/india%E2%80%99s-middle-class-big-enough
    http://www.cgdev.org/blog/india%E2%80%99s-4-10s-new-not-poor-not-middle-class-and-its-implications

    What do you think of those numbers? I would really love some discussion on this subject from fellow pundits. It seems unavoidable that India will continue to have one of the largest numbers of abjectly destitute people in the world but how many people will it have that are globally competitive and are able to attain a modern standard of living with modern aspirations? Will the size of this class be large enough to contribute to global knowledge and culture and establish India as a strong, influential region in the future multi-polar world?

    ReplyDelete
    Replies
    1. 2 reasons Birdsall data is an underestimate:

      1. Birdsall have used 66th NSSO household consumer expenditure data for her study. There is already a major controversy surrounding a more recent NSSO consumer expenditure dataset for under-reporting consumption by half-
      http://www.thehindu.com/news/national/consumption-data-contradict-high-poverty-estimates/article5979481.ece?homepage=true
      Besides, using Income as a parameter will always underestimates middle class size in India. This is because unlike West, more than 80% of India's workforce (and atleast half of GDP) is in unorganised sector- these include not just the poor but entire classes of traders, shopkeepers, middlemen and so on. For most (if not all) of this section, There exists no tax record or any other Govt record of their actual incomes. Obtaining monthly household consumption data is again a very cumbersome, unreliable and indirect affair. In my view, Fixed household assets and amenities [Accumulated Wealth that you can see] are a better (though still not exact) indicator of middle class status.

      2. That NSSO Survey was conducted 2009-2010-almost half a decade back. Given the fact Indian economy is changing VERY rapidly, its a big underestimate for 2014 even if it were exactly true for 2009-10 (Even 2011 Census data is significant underestimate for the reality of 2014). For instance, Mckinsey experts that I quoted , expected the size of middle class to grow from (their estimate of) 5% of population in 2007 to more than 40 per cent over the next 2 decades (nearly 20% by 2015).

      Delete
    2. Thanks for the reply and explanation of the likely underestimation in Birdsall's data. Would you or anyone else mind addressing the other points I brought up in my 2nd post below? Thanks!

      Delete
  2. Somewhat related(and I'm not sure if this should be tackled on another thread) is India's current education status and education policies that sadly indicate that the majority of its youth are not being imparted with the skills that they need in order to be competitive in the 21st century. PISA scores in Himachal Pradesh and Tamil Nadu in 2009 were abysmal. The two states only beat out Kyrgyzstan to attain the 2nd to last place among 74 countries. The results of the PISA indicate that Tamil Nadu/Himachal Pradesh's top 5% of students are 24 points behind the OECD average and nearly 100 points behind the AVERAGE student in Singapore/Hong Kong/South Korea. One issue that no one seems to have picked up on is that Kyrgyzstan also has a population that is 'east asian' in appearance yet scored even worse than the two Indian states tested. India withdrew from the PISA after this trial run and so it is the only data we have in terms of how India's students might compare to other countries in their learning levels.

    As I said though, what really matters is the absolute NUMBER of people who are globally competitive and in this case India should have quite a few in comparison to other countries due to its gigantic population especially if the economy and educational instruction is improved in the near future, correct? I don't believe that the PISA is an IQ test and rather the poor scores reflect the terrible education that these students are receiving(teacher absenteeism, student absenteeism, lack of teacher and student motivation,etc.) but still it does chafe me somewhat that if the PISA results are correct, India's top 5% is 80-100 points behind the average student in the east asian economies and 24 points behind the AVERAGE student in the USA. This can't actually be true, can it? The number of Indians/South Asians who are globally competitive would be paltry, if it were, no? To be honest it's affecting my confidence in my own abilities. I hate bringing this up but what should I do get rid of my own feelings of inadequacy after reading such reports? Apologize for the long post but would really appreciate replies! Thanks!

    ReplyDelete