Is there a middle class in Asia? Depends on how you define it


This page in:

A colleague from the Asian Development Bank visited the other day to talk about a study he is doing on Asia’s middle class.  Yet this is not an area we have focused on in the World Bank’s East Asia region – perhaps at our cost.  I quickly googled the topic and discovered a rapidly growing literature, including a paper each by Martin Ravallion and Nancy Birdsall just this year.  

Why this attention to the middle class?  Empirical evidence shows that growth of the middle class is associated with better governance, pro-growth reforms, even better infrastructure.  It appears that as people gain middle class status, accumulate savings, and acquire secondary and tertiary education, they are likely to use their greater political clout to press for accountable government.  This includes the rule of law, property rights (they now have more to protect!), and greater public goods supportive of growth including better infrastructure, education, fewer trade restrictions, and economic stability.  Interestingly, the larger the middle class a country has, the more likely it is to reduce poverty faster.  Ergo – if we are interested in growth and poverty reduction – we better start paying attention to the middle class.

But who are the middle class?  Alas, this is where the economics profession lives up to its billing (“ask ten economists a question and you get eleven answers”).

  • Banerjee and Duflo (2007)  suggest it should include those earning between $2 and $10 per person per day (in 2005 PPP terms – which will be the basis for all dollar numbers hereafter).  The bottom bound -- $2 a day – is the median of PPP (Purchasing Power Parity)-based national poverty lines of developing countries; those above this international poverty line can arguably be considered middle class in the limited sense that being middle class means not being poor.
  • Ravallion (2009)  goes further, anointing the middle class as those earning between $2 and $13 per person per day. Ravallion’s upper bound of $13 a day is derived from the US poverty line in 2005 (about $19,000 for a family of four).   According to him, the middle class would not be deemed poor by the standards of their own (developing) country, but would still be poor by the standard of rich countries (read:  the US).  This is an interesting proposition, but it’s not obvious why a rich country’s poverty line would define a developing country’s middle class. 
  • Birdsall (2010)  includes those earning more than $10 a day, but excludes those in the top 5 percent of the income distribution who she considers rich.   Unfortunately, Birdsall’s definition is so restrictive that she finds virtually no middle class in South Asia and very few in East Asia!
  • Yet others define the middle class as those within some range of the income distribution – for example, those with incomes between 75 percent and 125 percent of the median of the income distribution.   This definition has the obvious advantage of taking the concept very literally by carving out the middle of the income distribution – but falls prey to the obvious criticism that no matter how poor a country is, it would still have a middle class.

And there are others as well.

So where does that leave us?  Is there a common definition out there for the “middle class”?  What level of definitional purity should we seek – one common definition for all developing countries (similar to the Banerjee-Duflo or Ravallion definition) or different definitions based on country characteristics?  Let’s hear your views before I emerge with some thoughts on my own.

Join the Conversation

Matthijs van den Broek
June 02, 2010

A middle-class is defined by its free disposable income and its purchasing power. For example, the middle-class in Indonesia, where the salaries are low and the costs of living high, is something entirely different than for expample the middle-class in, let's say, Thailand, where salaries are higher but costs of living substantially lower. Therefore, in a country like Thailand, there isa true middle-class, whereas in Indonesia there is virtually no middle-class at all.

June 02, 2010

Maybe we should look for the answer to your question "who are the middle class?" beyond the world of economists and ask sociologists and anthropologists. I researched this question 15 years ago in India and found a wealth of literature! Frankly, I do not think we can define a middle class by the income level only; we need to look at consumption patterns, aspirations, values.

June 02, 2010

It is an intresting question to debate. Income is one indicator to define poor, middle class or well off. In my view, in addition to income,there are other important indicators to consider to define a class such as level of education/access to education, living conditions, access to information, decision making power, status in the society including rights, political awareness/power, etc.

Bakampa Brian Baryaguma
June 04, 2010

I think the middle class in any society comprises those people in charge of decision making. Now, "decision making" or "decision makers" if you like, is quite ambiguous. Precisely put, I mean the people placed in such positions as to be able to influence the economic and political dynamics.

The determinant here is one's purchasing power as stated by Matthijs van den Broek above. Once one is able to consume to his substantial satisfaction (i.e meeting all life's basics) then I think they qualify to fall among the middle class.

But this is not entirely all--for there is the question of status in society. But this is easily answered by one's ability to purchase and consume goods and services, hence the nexus between economics and politics.

Knowledge and access to information determine status and this is easily attained once one is able to purchase commodities.

June 04, 2010

It coud be that " the growth of middle class is associated with better governance , progrowth reforms, even better infrastrucure " may have a slight adjustment when it is applied to Southern Europe . Just imagining the apply to Asian scale may , eventually, sugest that it is a bit unrealistic as a concept for sustainable and inbalanced development . Great transitions may need some little adjustment of concepts to work out in good the results of course.

Muhammad Owais
June 05, 2010

In my opinion, it has nothing to do with the numerical value of dollars earned per day.

The better way to define: It is the capacity to spend well above what is required for basic necessities such as food, education, health and shelter.

The resultant savings will then contribute towards the development.

punya rijal
June 06, 2010

It is diffcult to fineout the middle class just on income.In least developed country like nepal,it should be base on consumption.Having a car but less then 1200cc can be definied as middle class .

June 06, 2010

Determining middle class using purchasing power parity will have different results from country to country and from conditions to conditions (ie, economic crisis).

I personally think, that despite all definitions, middle class is when a family, as a whole, can save a portion of their income after all their normal expenditures (basic + minimum requirements ie. going to movies, eating out) being deducted.

As long as a family can not save for their future expenditure, they are not middle class. They are still considered lower class despite their earning is more than USD2 per person per day.

When saving is practised by a family, then they can be considered middle class, which by default shows some kind of governance in the family. This is a micro governance shown by a middle class. Translate this into community, we will have an accumulated savings and collective governance.

June 08, 2010

I think national budget is better determinant of middle class, since it shows the money in compare to money out, and how large the deficit or surplus if found. If the deficit is so high, let's say 15-20 percent to the GDP, IT IS WISHFUL THINKING TO DISCUSS WHETHER THRERE is a middle class in that country.
Access to health and education are very significant variables.
Demographic status, hether the country is youthful, transitional, mature and aging are to beconsidered also.
I would say that most social issues and phenomenon are place base rather than people base, and this applied to middle class within the country.

June 29, 2010

I think Birdsall has it more or less right in setting the lower bound at PPP10 a day, and for the reasons she sets out. However, more variation might be allowed for her upper bound, depending on the mean PPP income of the country in question. For countries with higher mean incomes, the upper bound might usefully be set at the 90th percentile, whereas for those with lower mean incomes, it could be the 95th or even the 99th percentile. But it should be income, not consumption, based. If it is to be consumption based, then the lower bound should be moved down.

Banerjee & Duflo and Ravallion have it quite wrong to set the lower bound at PPP2, mistaking what is arguably a useful yardstick for measuring (the progress in) poverty (reduction) at the world level for a realistic indicator of a standard of living. I would be willing to bet anything that neither of them could have anything that might be considered a middle class existence on PPP4 a day in any of the middle income countries of East Asia.

That Birdsall's measure results in no middle class in South Asia: doesn't that capture the view that South Asia, or more specifically, India's, growth of the past couple of decades has not had as much spread effects as some might like to believe? Comparing her table 3 and 7 is also instructive. While urban India apparently has no middle class by her criteria, if we were to shift her upper bound up to the 99th percentile, then it would be 4% which, in absolute numbers, would be sufficient to support the malls sprouting up.

June 30, 2010

Dear KJ

I really like your point -- but have the following thoughts. First, remember that these are per capita numbers -- so assuming a typical household size of say 5 (a recent survey found it to be around 4.8 in India), even a PPP$2 a day translates to a household income of PPP$10 a day -- which, using PPP exchange rates, translates to about Rs. 150 a day in a country like India (Rs. 4500 a month), and one could conceivably argue that this may be just enough to provide a few creature comforts that may, at a pinch, be described to as the lower end of the "lower middle class".

On the other hand, applying your (and Birdsall's) PPP$10 a day would mean a PPP$50 a day income for the household -- or around PPP$18,000 a month -- a very decent sum by Indian standards (over Rs. 270,000 a year!).

Second,the word "middle" in "middle class", suggests that there should be some part of the income distribution above and below this group. By moving the upper bound of the middle class to the 99th percentile, you are reducing the portion above the middle class to a wafer-thin 1 percent. True, in India, that may make a lot of sense -- because an upper class of 1 percent actually represents a resounding 10 million people! And as you point out, the middle class according to this definition would be a huge 40 million. Incidentally, I did find some data on India that is instructive:

PPP$20 and above -- 8 million
PPP$10-20 -- 15 million
PPP$ 4-10 -- 74 million
PPP$ 2-4 -- 170 million

Total -- 267 million

This is where the "great Indian middle class" concept comes from -- the basis for the usually bandied about number of 250 million that one sees in newspapers and magazine articles.

Best regards


Alexander Jett
February 09, 2011

I wholly agree that this or as another commentator said, the purchasing power of the family should be the basis. I think most would agree that a middle class family typically is living beyond subsistence and owns some property (not necessarily land). The dollar amount of income to accomplish this is different in all countries.

March 02, 2011

You may wish to look for the people who have some form of capital rather than just check their revenue. I like the approaches that combine human capital (e.g. education level, socio-economic category) and economic capital (some household equipment goods, maybe land / housing, insurances/pensions).

March 05, 2011

I would insist that social stratification systems are far more complex than mere income categorizations. Not only financial capital but also social capital (positioning within a social network) and human capital (education, health) influence an individual's abilitiy to access resources and opportunities. Taking income alone could be misleading since a country's income distribution might be dependent on very different social structures thant those of another country. Altough outcomes are certainly comparable (earnings per day for example), such comparisons may hide important structural differences between societies. A $3 earner in Malawi may have very different capabilities than a $3 dollar earner in Syria or Cuba.

Hence, I feel the criteria to categorize the middle-classes should be tailored to the type of research that is to be pursued (provided that the limitations of such categorizations are understood and duly noted). Cross-country comparisons would of course demand a much simpler method (perhaps related to the aforementioned three), whereas case studies may be able to afford the luxury of a more holistic method of a higher level complexity (where capabilities of different sort, not only income, can be taken into account).

Vikram Nehru
March 08, 2011


I like (and agree with) your point that what you measure depends on what question you want answered. Single country longitudinal studies may well want to develop a composite measure of flows (income) and stocks (capital) to identify the middle class, although doing this rigorously presents interesting conceptual challenges. I guess my original blog entry focused on attempting to get comparable measures of the middle class across countries for two reasons: first, to place alongside other economic indicators to compare where countries were along their development paths; and (b) to aggregate the numbers of the middle class across groups of countries to see how they may stack up across countries in other regions or at other income levels.

As you point out, getting comparable data across countries is challenging at best (and a fool's quest at worst!), so a simpler measurement is warranted for such purposes. Trying to include capital stock measurements in cross country analysis opens up a can of worms. A long time ago (1994 to be precise), I worked on a research project to develop a comparable measure of human capital across countries (its a 1995 JDE article) and I know the shortcomings of these methodologies and how hard such measures are to compute.

Best regards


Ramesh Kumar Nanjundaiya
April 04, 2011

This was a response to an earlier article by James Seward (please see below). I feel this article holds good for the present discussion as well. Per my personal observation the growing Indian Middle Class can be generally classified on their "buying power" - which effectively is their monthly earning and/or take home salary. The range would be a monthly salary or "earning band" of INR 20,000 to 35,000 or approx average USD 700/- per month. I personally feel this "class of people" number almost 300 million today (comprising of urban and rural). This is certainly a big and a growing group which actually spends thus creating a huge and a growing market. This phenomenon has attracted all types of Multinationals to the Indian market space along with a growing consumerism off take particularly since the last 6 years and Indian companies/services are also enjoying the boom.

Is East Asia leading the world out of the crisis?
James Seward has summed up very well on the current economic situation of the seven countries in this region. He has made a useful observation on the equity markets in these countries to draw some sort of conclusion to see if indeed the equity markets are a leading indicator of economic recovery. While it is true that the major stock markets in the region have regained all of the losses of the 2008 and that they are up by over 66%, with Indonesia, Vietnam, Thailand and Korea leading the pack. Is this the true indicator - it could well be. As a comparison, if one were to look into the Indian economy in the same period that James has analysed the above markets, one observes that the Indian Economy is simply booming. The reasons are varied, but one which has really propelled up the Indian economy (in the last 6 years) is the growing buying power of people in the so called "Middle Income Group" which in the case of India represents almost 400 million people. This is a huge market to cater to and is growing. This group is the one which is pushing demand locally and thus giving a boast to the economy. It is a life cycle change in the population group. The same phenomena is happening in the seven countries of the East Asia analysed above but with different degrees. The Middle Income Group of people in these countries (varies from country to country) is driving demand and thus helping the economy to grow. This is the group which is spending on all goods and related services. Because of such a growth demand for goods/services, banks will witness increase their lending in these countries in the next couple of years. Once that happens, the economy grows, when this happens, the equity markets become much more active and again the economy grows with more people coming into the "Middle Income Group of People" or the people with buying power or cash to spend.

July 16, 2012

I think it is difficult to define the "middle class" based on the income or consumption. For example, an ordinary family in East China can be regard as the upper class in the West China, if we define different classes only based on the income or consumption. As far as I am concerned, classes can never be figure out very clearlly.

Sanjeev Ahluwalia
July 10, 2012

The middle class should be relatively easy to identify is we agree that it has the following minimum characteristics: (1) Income levels which permit a degree of capital accumulation sufficient for acquisition of consumer durables and a house (2) Disproportionate use of disposable income for education of children (3) Predominanatly employees or professionals (accountants, lawyers) rather than entrepreneurs (4) Urban dwellers.

These characteristics account for most of the economic and politcal incentives which drive this group (eg. tax breaks on investment in a self occupied house, lower taxes on salaried income, tax breaks or state subsidies for educational institutions, state subsidy for mass housing etc, state subsidy on urban metros, meritocracy in state recruitment etc.)