Was the Indian caste system efficient?

A new paper looks at some of the efficiency properties of castes:

The caste system in India has been dated to approximately 1000 B.C. and still affects the lives of a billion people in South Asia. The persistence of this system of social stratification for 3000 years of changing economic and social environments is puzzling. This paper formalizes a model of the caste system to better understand the institution and the reasons for its persistence. It argues that the caste system provided a tool for contract enforcement and facilitated trade in services, giving an economic reason for its persistence. A caste is modeled as an information-sharing institution, which enforces collective action. Trade is modeled as a version of the one-sided prisoner’s dilemma game, where the consumer has an opportunity to default. Consumers who default on a member of a caste are punished by denying them services produced in the caste. Various features of the caste system like occupational specialization by caste, a purity scale, and a hierarchy of castes are shown to be equilibrium outcomes that improve the efficiency of contract enforcement. The implications of the model are tested empirically using unique census data from Cochin (1875), Tirunelveli (1823) and Mysore (1941).

In other words, other caste members enforce norms on you and if you don’t follow them you are kicked out and you cannot easily join another caste.  Sounds like my idea of fun.  I have a few points:

1. No way should this paper spend so much time on a formal model.

2. The tests proffered on p.36 are related only tangentially to the paper’s main propositions.

3. When it comes to normative issues, the author can do no better than to write: "This should not be interpreted as saying that the case system was free of inefficiencies."  And that comes only on p.46.  Ha!

4. The paper commits the fallacies of excess functionalism.

5. Virtually any destructive institution which keeps economic transactions on a smaller scale may make contract enforcement "easier" in some regards.

6. This is nonetheless interesting work, and many more people should do research on this and related topics.  But in terms of emphasis this paper is way off base.

The pointer is from New Economist blog, which offers related links.  Readers, are any of you willing to defend the caste system, if only in part, on economic grounds?


I'm not, but Marginal Revolution is full of commenters who approve of inequality. Their economic justification is that it encourages people to work harder. They'll also say that low caste people shouldn't be so concerned about being low status and what others think about them and that they should instead work on trying to make peace with themselves.

Is there a difference between class and caste? Only one of degree. Humans naturally self-segregate. It's equality that is unnatural.
Admittedly general thoughts that don't address the contract question.

Suppose people take reincarnation seriously, and believe that prosocial behavior in one life leads to being born into a higher social ranking in the next life. In that case, a caste system with low levels of social mobility raises the marginal rewards for good behavior in this life. The greater the caste stratification, the more people will try to make long-term social contributions.

I'd like to hear what Greg Clark has to say about this.

Efficieny is by all means relative, so is a caste system "efficient," well that depends on your benchmark. Is it efficient compared to a free market, emancipated populace, I think we see the answer is clearly no. India is no where comparable to the United States, Great Britain, France, etc. in terms of its economic development, and that's very likely due to economic choice models.

Is the caste system efficient compared to despotism, with constantly warring factions vying for "king," yeah probably, because as mentioned earlier with the aid of religion helps to limit competition for the top "caste."

As it relative to relative efficiency 10 compared to 1 is a big number, 10 compared to 100 is still small.

Dirk thinks income inequality and caste are the same or similar phenomena. That informs us on Dirk's qualities as an economist. Maybe he should take lessons with Peter on the subject.

Back to basic economics people. Prosperity comes from productivity. Any system that prevents any group of persons from maximizing their personal productivity hurts all of us world wide. How many cures for cancer may have been lost to dalits designation to menial jobs?

I can't believe any economist would waste time exploring the efficiency of anything that is so obviously inefficient on its face.

I can't believe any economist would waste time exploring the efficiency of anything that is so obviously inefficient on its face.

Chalk it up to the desperate search for counterintuitive results.

I cannot & will not justify the caste system on moral or economic grounds. The only argument in favour of it's perpetuation is that sited for Russian communism or South African aparthied - that dismantling them would (did) create short-term upheaval & conflicts. Nevertheless, all three systems were wrong and deserve(d) to go.
That said, where the infrastructure of a country is so poor, vested interests so powerful, and corruption as rife as it is in India, there is no guarantee that change will be an immediate improvement. That's not an argument for the status-quo, but a cause for caution when implementing much-needed change. The tribulations of Russia's economic liberalisation teaches us that doing the right thing 'too quickly' can be sub-optimal in the wrong circumstances. All the Indian 'affirmative action' legislation I've seen proposed is too one-dimentional & clumsy to make a real difference in the right direction.
There's a lively & fascinating debate in India at the moment about the nature & quality of leadership required in such a vast, disparate & rapidly evolving nation (s.f. the Time of India's 'Lead India' campaign) - I wish them well in their quest to find the answers to the conundrum, and continue to watch with interest.
PS I basically agree with what Tom Kelly says in the post above, but still believe that it's worthwhile to study the efficiency of such a clearly inefficient system. The system (of castes) has survived in the face of economic reality, and it's important to understand why such anomalies persist - there must be an underlying reason.

@Joel B: be very careful about making such direct comparisons.. Remember that India has been ruled by foreign powers for hundreds of years.. what has been the economic loss of foreigners leeching on your countrys resources? and who has benefited?

notspeaky: my point is exactly that it have been efficient.. but perhaps it no longer is.. just as arranged marriage has been efficient.. and in fact this is closely related to the caste system.(I do believe it is no longer efficient and perhaps haven't been for at least some hundred years or so..)

Remember that not many hundreds of years ago the "western" world had very similar systems in place (if not worse and more centralised systems)...

If you wonder why the caste system still exist I would look at the indian people being protectionistic about their own culture for many hundreds of years as they were under threat of foreign cultures and religions. thus most indians are quite conservative and sticks to tradition.

Why is it so counterintuitive that inefficient systems can exist for thousands of years? Have people lived in a marketplace of social structures where social structures that don't maximize global efficiency are doomed to collapse due to the nature of the competition? Huh? Exactly.

If you wanted to understand the longevity of the caste system you would have to look at the margins of social systems, see what causes them to change and what allows them to perpetuate themselves.

Nice. I'm enjoying the mean and mad Tyler.

Many people covert to budism to exit the cast system.The risk their life doing that, their IQ is not low then

Humans in Australia did not naturally self-segregate.

@peter: The caste system is very much in force today in India - even urban India. It is tabu - though few want it to be abolished.

Care to justify that? Or did you mean "a few" instead of "few"?

The ground reality in urban india is weighted more towards economic class rather than "caste". I am of a very "low" caste by birth, and make a good living in the IT sector. My maid is of a much "higher" caste by birth. I'm sure that had something to do with the fact that my parents were middle class while hers were much poorer.

Once you cross a basic economic threshold, which takes a generation or two of hard work and a bit of luck, opportunities are relatively equal in urban India for the low castes.

Dalits today suffer from the fact that they start off much poorer and with much less of a family history of education than the upper castes.
From whatever anecdotal evidence I can gather from observing friends, family and people around, there doesn't seem to be a significant difference in basic ability among castes, only differential access to education and the knowledge of how to use education to one's economic advantage.

Perhaps the rational economic solution would be a lump-sum redistribution of wealth across caste lines?

R. D. Sack's theory of territoriality is also a well regarded functionalist theory as to why the caste system might have become instituted without the need for strict coercion.
Don't see how it can't be justified any more. There's far more efficient ways to deal with exogenous shocks to the society - *ahem*trade*ahem*, *ahem*capital*ahem*.

but one cannot change one's class no matter how much drooling, name dropping and sycophantic behavior a (usually middle class person) does.

This is an overstatement. It's not a matter of name-dropping and kissing ass; you move class through education and employment. It's certainly possible, if somewhat unlikely, for someone in America to change classes.

before u jump the gun...there's no real evidence that lower castes score less on IQ tests than upper castes. That is just assumed due to the general correlation between socioeconomic level and iq, and the correlation between caste and socioeconomic level.

before u jump the gun...there's no real evidence that lower castes score less on IQ tests than upper castes. That is just assumed due to the general correlation between socioeconomic level and iq, and the correlation between caste and socioeconomic level.

Caste is really about 200 years old.

The beginnings of the caste system are CLEARLY visible already in Bhagavad-Gita as well as The Sacred Laws of the Ayrans Part 1 and 2

It was already a firm part of Indian life by the writing of The Laws of Manu - between 200BC and 200AD.

Yeah, it changed somewhat after the british arrived but the core notion of the caste system - your family is your job is your caste - IS thousands of years old.

You really believe that the single quote you offered demolishes Dirks' argument in "Castes of Mind"!


"1. The great sages approached Manu, who was seated with a collected mind, and, having duly worshipped him, spoke as follows:

2. 'Deign, divine one, to declare to us precisely and in due order the sacred laws of each of the (four chief) castes (varna) and of the intermediate ones." - Laws of Manu chapter 1.

"(There are) four castes--Brâhmanas, Kshatriyas, Vaisyas, and Sûdras." - Laws of the Ayras Chapter 1.

Learn something dumbass.

No one is dispute the fact that there was a concept of caste for thousands of years on the Indian subcontinent. One cannot quote an ancient document, and then infer from that document that caste, as we know it today, was the same back then as it is today. Moreover, the concept of caste was nearly absent from the northern area of the subcontinent. Caste was strongest in the southwestern region of India. One cannot infer from ancient texts how an entire civilization lived, ruled, governed, and structured their society.

As for r4d20, who wrote: "You can believe Dirk all you want though. I mean, why believe contemporary Indian sources when you can believe a White European who wrote hundreds of years after the fac?."

Well, I'm going with the evidence, and Dirks presents the most compelling and tightly argued case that I, or the entire profession of historians of colonial India, has seen.


The notion that "your family is your job is your caste" existed for thousands of years is under attack in the academy. I am not an expert in the subject, and the subject is dauntingly complex, but I do know that the recent scholarship has worked away from the view that caste was the primary factor of self-identification in precolonial India.

But what do I know. I'm just a 'dumbass'.

I do know that the recent scholarship has worked away from the view that caste was the primary factor of self-identification in precolonial India.

But what do I know. I'm just a 'dumbass'.

1. It looked to me as if you were dismissing my point without even looking for a second at the sources I cited - and since the sources deals largely with the exact rules for each caste, it was kind of annoying to have them be ignored in favor of a single book by a single author.

2. Part of my reaction was because the subject of Indian History is the subject of a lot of pseudo-historical bullshit and propaganda from Hindutva advocates who, for political reasons, blame everything "unpopular" in Indian culture on the British while claiming that everything good about their culture is native to India. They are as annoying to real scholars as ID advocates are.

Yeah the caste system really doesn’t make me think happy thoughts when it come to economics. And just because the caste system “works† should not mean it was the best way to do things. There are better ways to do something. At least when it comes to our system you have opportunity to adavance above your class, to better yourself and that gives you a goal. If you cant better yourself why try?

Is it just me, or is the argument about purity bizarre?

How does refusing to produce goods for myself improve my bargaining position? Would I really get a better deal from restaurants if I committed to never cooking for myself?

I know there are some cases where reducing your options can increase your bargaining power. Eg the military commander who burns the bridges behind him, ensuring that his troops have nowhere to run to. But I don't see how refusing to produce something for yourself can increase your bargaining power.

And of course trade manages to happen perfectly well without purity conceptions. I don't know anyone who thinks in NZ that cooking for yourself makes you impure, but restaurants still do a thriving business.

The ultra-Orthodox Jews are a different case as they are producing club goods.

DAS reaffirms my point concerning class. Having an MBA, wearing nice suits, buying a mansion, sporting jewels, sending kids to expensive schools, etc., does not your social class make. Consider these people: Clinton, Reagan, Bush and Kerry. Consider their family backgrounds, moving back to the 19th century (which is nothing compared to Europe.) Look at how long and how broadly these *families* are connected to wealth and powerful positions. Look at what families they marry into, and look, if possible, at who these families exclude. Earning a million dollars a year and getting an education does not change one's class. It changes one's income. It changes one's buying power. It might change the odds that one's grandchildren might be considered part of a 'higher' or more exclusive class, but a person will always be what they are. If there is doubt about this, consider the your own reaction when someone you consider to be from a lower class buys the house next door.

The caste system is economically efficient? Not aware of such things. I am an actual untouchable. Let me tell you that the untouchables and lower caste people used to ( some say they still are in some areas) to eat the wheat grain cleaned out of cow dung. You may find it difficult to believe. High caste Hindus would deny it . But the point here is economic efficiency. Such a condition of untouchables does not exactly point to efficiency of any kind as such. It points only to crass exploitation. Secondly which kind of society needs 20% of its population to do the unclean jobs? Certainly no society creates demand for unclean goods and services that it needs every fifth person to provide them. Thus it is a case of artificial excess supply of labor. The quantum of this excess artificial supply of labor (through heredity) makes the Hindu caste system extremely inefficient.

A couple of points-
1) India was a mixed regime economy with a metropolitan cash nexus parallel to a system of non-monetized specialisation & division of labour where agicultural land and surpluses were periodically redistributed according to some ill-defined communal collective bargaining which more often than not meant whole sub-castes (endogamous occupational groups called jatis) voting with their feet. During exogenous monetary shocks/ natural disasters the unmonetized jati system of specialisation and exchange was the default value. It kept things ticking over and allowed the clearing of new land and rural settlement with low capital investment but the full complement of services. Theoretically, endogamy was supposed to facillitate the diffusion of new techniques of production and thus raise productivity. Religion tends to say what is is right. God wants it that way. Vedic Hindu religion was particularly suitable for this because it emphasised the equality of all paths to the deity- i.e. the notion that utterly opposite codes of behaviour (customary morality) were equally valid. This lead to the notion that every occupational jati was engaged in an imitatio dei- i.e. the potter feels God is a potter & hence derives a psychic satisfaction from making pots, the thief reckons God is the ultimate thief etc. Hence each jati reckons it is supreme as being engaged in God's quintessential activity. Each jati develops its own spirituality and coercive system.
2) The metropolitan cash nexus, when supported by State power used for extraction of surpluses, does not in its expansion dissolve the default system, but leeches resources of it causing it to crash. This is because the initial rise in productivity incident on monetisation is ultimately swamped by ecological degradation arising from collapse of public good provision. The metropolitan cash nexus esperiences a tulip bust and learns fear of a more fundamental collapse affectiong every aspec of Socialisation. Fear of the crash means the metropolitan cash nexus ultimately resurrects jati as varna (legitimating ideology for social stratification).
3) Only if the State becomes effective in provision of public goods at the village level can the crashing default system be expunged from the directory. However, if the State is dependent on its survival on the crisis of the default system- i.e. it is a protest against what it perpetuates- this aint gonna happen.
4) Though jati type caste appears to be, under ideal conditions, a co-operative solution that would dominate competitive solutions- giving rise to dreams of 'RAmrajya'. Gandhian village swaraj, Buddhist Socialism etc- this is just a pipe dream. It is not productivity that rises but the amount of time people spend in 'consciousness raising' and other such magical practices.
5) Caste in present day India is about people from larger jatis taking power by claiming to speak for smaller jatis that have, arbitrarily, been adjudged to be of equally low status. It represents a redistribution of wealth from the poorest to the richest in these arbitrarily defined 'castes'. This is not done in a Pareto efficient way. Rather there is a huge and growing deadweight cost to the whole economy. The rich can escape this, because, in India, increased life chances equals increased elasticity of supply and demand (i.e. the rich can always get what they want by circumventing the system. Not just a Black Economy, India has always had (infinitely many!) Anti-States, Anti-Commons, Anti-Relgions (as well as Anti-Anti-States etc) membership of which is the path to thrive.
5) Some social customs can (only if everyone is very good in the first place)internalise externalities, improve Schelling focal point choices, turn prisoner dillemmas into positive sum games etc, etc. But why bring caste into it? The English word caste dervies from the Portugese and Spanish notion that degree of miscegenation determines 'nature' and hence ought to determine social status. Does anyone believe that sort of racial nonsense anymore?
6) One good point about jati-dharma (caste based religion) is that it recognises that from the moral point of view there is a huge range of choices which are equally acceptable. However,some modern Hindus believe that their own vegetarian jati-dharma is the universal morality and seek to convert everybody to an irrational and socially very costly (and potentially environmentally disasterous)code of conduct. Same point applies to Islam, missionary Christianity, Eco-feminism, Marxism, Hitler worship etc.
Caste is crap. However economic analysis of its causes and consequences should continue so as to prevent history books from spouting silly conspiracy theories- the Brahmins did it! No, it was the British. Actually, you're both wrong- it was the lizard people from Planet X.
nuff said

Caste system stinks! That's wrong to lable people. Just like high school cliche's. It's all bogus

I, Joey, think the caste system is amazing!!!!! guess what?!??!?!!?!?.....i am a cow now because i died!! because I lost a hockey game, and i didnt score): oh well i still think im a beast!!!!!! so bck to the catse system....its awsome(: and again...im a cow...i should make a song up...Imma cow Imma cow imma imma imma cow!!!! hahah byeee

lower cast people who r living in higher socioeconomic class and having benefit of reservation since there grand parents, and still they have all facility of good education but needs reservation for professional education and jobs on the other hand poor upper cast has no social support to get education and jobs. lower cast student can be a doctoer even if he is not having iq to make 50% marks in school but a 95% marks holder can not be a doctor because for 50% of medical seasts are already reserved for lower cast people irrespective of there economic status....

I recently came across your blog and have been reading along. I think I will leave my first comment.
I don’t know what to say except that I have enjoyed reading. Nice blog.
I will keep visiting this blog very often.
By the way,
Blogger how when you visit my blog,
My Blog have been created for the satisfaction of consumer of all.

Comments for this post are closed