Syndicate content

The Business of Knowledge

Kaushik Basu's picture

A large part of the task of economic development in the world can be achieved by carrying existing knowledge from where it is available to where it is not. The creation of new knowledge is of course important, but when one looks around at the large areas of unwarranted darkness in the world, it becomes evident that there is a lot to be gained simply by knowledge arbitrage. But the reason why this does not happen, large knowledge gaps persist, and we fail to deliver even when we have the knowhow is that knowledge arbitrage is not as easy as it may appear at first sight.

We have the knowledge needed to eradicate polio from the face of the earth. Years of research gave us the vaccine, first in injectable form and later as oral medicine. By 1962 this was licensed. Yet even now well over a thousand children contract polio each year. This is the reason why we are shocked when we get news of nurses and doctors participating in vaccination campaigns being killed. The most recent was the case of nine women killed in Nigeria by gunmen suspected to be part of a radical Islamist sect. Similar incidents have occurred in Pakistan and Afghanistan. And there is no getting away from the fact that, in many places, terrorists succeed in carrying out these attacks because of pre-existing local suspicion about the polio vaccine.

Many social scientists express shock and a failure to understand such ‘irrational’ resistance against a proven drug. It always troubles me when that happens. It is one thing not to condone, quite another not to understand. Let me try to help them understand with an imaginary scenario. 

In North Korea, a team of scientists under central  government supervision develops an oral vaccine that enhances the intelligence of children. Years of research in Pyongyang show that the medicine works wonders and has no side effect. This mystery drug that, despite the best effort of North Korean researchers, has a strange purple color, arrives in American drug stores with pamphlets endorsed by North Korea’s leader  urging all responsible parents to administer this to their children at the earliest opportunity.

We can be sure that parents will not queue up for this drug. Indeed it will be viewed with great suspicion. Stories in Pyongyang Daily News of how the country’s leader told his wife that politics must not stand in the way of the happiness of children and he is going to spread this drug to all nations, beginning with the United States, will be dismissed as propaganda.

It is easy for us to understand this suspicion. It should be no harder for us to understand the suspicion about polio drugs in societies that are completely isolated from the modern world.

What this illustrates is that knowledge products are likely to be afflicted by what economists, following George Akerlof’s seminal paper, refer to as the ‘lemons problem’. The used car or the lemons market, Akerlof argued, functions poorly because the seller has disproportionately greater knowledge of the product being sold than the buyer. It will be immediately clear that knowledge products will have this problem of ‘asymmetric information’ even more seriously because, by definition, the developer of the knowledge product will know more about the product than the customer. This means that the seller of this product can take advantage of her superior information and, as a consequence, the buyer will be plagued by concerns that the seller will smuggle in a Trojan horse under the garb of a new product which the buyer is not in a position to evaluate. This is the fundamental reason why knowledge arbitrage has worked so poorly.

In exercises of knowledge arbitrage, the asymmetric information problem manifests itself in two different ways. The first is the conventional one in which the seller can pass-off as knowledge what is not so simply by virtue of the seller’s greater information. The buyer may not be able to tell which knowledge product embodies real knowledge and which does not but knowing that a certain percentage of such products may indeed be duds, the buyer may hesitate to enter the market in the first place. This is the classic Akerlof problem that showed why the used car market functions so poorly.

The second problem arises from the fact that the supplier of knowledge may have an interest which is at variance with what the demander (for instance, ordinary citizen) wants. Suppose the citizen wants a financial product that protects him from risk, and the supplier who is a moneylender knows of a complex financial product that leads buyers to penury and enhances the wealth of the moneylenders. It would not be surprising if the supplier tried to pass off such a product as a good hedge against risk. Knowing that this is possible will make the ordinary citizen hesitant in accepting such a product. This is what makes the problem of knowledge transmission so difficult.  
The World Bank, which is engaged so critically in the knowledge business has to contend with this problem every day. What we are only now beginning to understand is that there is more intricacy to the art of knowledge transfer than most people, including trained social scientists, had realized. We have to do a lot more research on this but one thing is clear—much will depend on how we can bridge the trust deficit that exists between societies and peoples, between those who understand the knowledge product and those who do not, but need it.

In the late 1980s and early 1990s I used to go to a remote village, Nawadih, in Bihar, now part of the newly formed state of Jharkhand, to do research. My self-confidence that I easily relate to people and they can see that I am trustworthy received a rude shock. I just could not get the villagers to open up to me. They could not be sure that I was not a birth-control agent from government who had descended on them with some sinister plan. What finally broke the ice was an intermediary—an English speaking villager, whom I had known from years ago. He trusted me and the villagers trusted him and in these two steps we managed to bridge the chasm of mistrust at least partially. With him on my side, occasionally adding a sentence or two and correcting my grammatically flawed Hindi, the equation with the villagers changed altogether. They spoke freely and I began to understand many of their problems, pertaining to land markets, tenancy and usury that I had gone to study.

There is so much malnutrition, morbidity and illiteracy in the world that we need to do fundamental research on foods that are nutritious, on inventing drugs that can prevent or cure disease and techniques that can help children learn better and faster. But it is also true that carving some time out of such fundamental research to understanding the art of knowledge arbitrage can yield greater dividends than most people realize.


Submitted by Anonymous on
The Bank produces a lot of research that is not operationally relevant. Often, I read papers that add no value….so much so, that I now hesitate in reading a paper without a referral, fearing it will be a waste of time. Can the Bank rate research papers? Or even more so, have a mini-Oscar ceremony for operationally relevant research papers?

Submitted by S. Akhtar Mahmood on
Thanks to Prof. Basu for highlighting this important issue. Just one reaction. Bank staff producing analytic reports are usually expected to come up with policy recommendations. Indeed, their reports may not be cleared without this. However, I have often seen the authors of analytic reports struggle to come up with policy recommendations, not because they are unable to think policy or their work was not of good quality but because sometimes the analytic work does not immediately yield policy prescriptions. As a result, we often see half-baked policy recommendations, or recommendations couched in very general terms. These do not enjoy credibility with our clients and other stakeholders. Moreover, once they become skeptical about one part of the work, i.e., the policy recommendations, some people may become cynical about the rest of the report as well even though the rest may be of high quality. A better approach will be to focus only on a good analysis and then have a patient process of jointly working with clients to come up with policy solutions based on the analysis, and augmented by any other knowledge the clients may have. Such an approach will help build trust and enhance knowledge arbitrage much more than a one-shot, high-profile dissemination event with ministers. S. Akhtar Mahmood Investment Climate Department. WBG

Submitted by Piyush on
The write up very succinctly highlights the issues of trust based on credibility of the surveyor and the endorsement it backs. In my defence, I d say I read this article only because it was Kaushik Basu and World Bank in the first place.

Submitted by Anonymous on
By highlighting the reality of the difficulties in the diffusion of innovations, Prof KB implicitly highlights the increasing role of (agricultural or medical or...) extension process in different parts of the world in the diffusion of innovations. Prof KB will know that 58% of the total factor productivity in India's green revolution was due to agricultural extension (the famous study by Evenson et al), the highest, even surpassing the role of Genetics and plant breeding which bred new seed varieties of rice and wheat. Unfortunately, the 'gram sevak' or 'village level worker (vlw)' who spearheaded India's green revolution has vanished from the State Department of Agriculture throughout Indian states, since the Governments have abandoned those positions, for the reasons best known to them. I once again suggest Prof Kaushik Basu to impress upon the GOI and the State Governments to bring back this position to enable our farmers and rural people to benefit from the VLWs who played samaritan role in the diffusion of information and innovation. Since the Developmental programs framed by different political parties are increasing to catch the attention of the political economy, and since (agricultural) technologies are also becoming increasingly scientific and intricate, it is desirable to have a Gram Sevak or VLW at least for a group of villages at Panchayath level. Thanks to Prof Basu for his excellent blog.

Submitted by Anonymous on
Is the Akerlof/"lemons"-explanation correct? If there is no financial transaction (assuming polio vaccines are distributed for free) and no quality differentiation involved, there cant be any "lemons problem", isnt it? The polio vaccine "market" in the cited countries does not work/collapse, because the lemons are driving out the goods, but because of either asymmetry of information in general or because vaccines are experience/credence goods...

Dear Prof. Basu While knowledge transfer is important, even bigger is the issue of knowledge formation and its connection with development. In India, we face a severe informalization of "practical" knowledge formation. Most formal structures of knowledge formation (universities, formal R&D etc.) serve global markets (IT, automotive, consultancy and so on) while local problems of water, energy languish. I think the WB can do a lot more to re-orient formal knowledge production in India. See paper below. Regards, Milind Sohoni, IIT Bombay Sohoni, Milind Ashok, Knowledge and Practice for India as a Developing Country (February 2, 2013). Available at SSRN: or

Submitted by Prashant A. Bhonsle on
Extremely relevant and well made points indeed! According to me the information asymmetry is as much (actually more as per me) a function of ‘basic design’ flaw of the system as it about barriers of language, culture and trust issues because of bad experiences of past. If the age old belief that no design is ever perfect and always work within the ‘right context’ is correct than is it too simplistic to understand that the context is ever dynamic and the system design will always have to play ‘catching up’ game to be relevant & effective? Classic example is that because of this stale or slow changing design of (either) democracies or/and autocracies, economics have become the heart of the system for most of human beings/countries rather than the values and purpose. In most of the ancient cultures, values, principles and larger purpose formed the core of all structures (societal, political, philosophical or even spiritual). Every decision was squarely based on pre-laid/agreed values and purpose. That singularly drove human development and evolution over thousands of years…till greed started taking over! We need 'Thought leadership' based on strong human values, principles and larger purpose.

Submitted by Sanjeev Ahluwalia on
Yes, the process adopted for knowledge diffusion and quality certification matter. The best way to market a water filter is for the seller to drink the water processed by the filter. At the heart of this problem in development is that people are right to distrust those who provide solutions for problems they themselves have never had to face. Does one have to be poor then to solve the problem of poverty. Jean Drez certainly thinks so but clearly, it is not the key prior. However, credibility as a poverty basher matters and credibility is partly about being able to share both the pain and the joy of those one seeks to benefit....empathy is a function of faith, sincerity of purpose and "quality time" spent in the environment we seek to change. The operational implications are that decentralization is key to operational effectiveness......but are we on the same page?

Submitted by Govind Hariharan on

I hadn't thought of the knowledge diffusion problem this way and hence was quite illuminating. Two issues I would like to bring up in this context: 1) the huge role played by divergent claims for example inthe safety of vaccines even in the scientific community. One wayto characterize this is in incorporating uncertainty. The FDA for instance did some studies on uncertainty in policy context that I thought can really help address this. With economic development problems, this issue of uncertainty may be even more prevalent and hence relevant. The lemons problem needs to be expanded upon with uncertainty and not just known probabilities. Lower uncertainty in quality and you increase trust. Glad to provide references if interested, 2) maybe warranties

Submitted by Dilip on

Thanks Kaushik. You highlight a key aspect of implementing the business of knowledge. A related strategic issue which the Bank (in aspiring to become a knowledge bank) must grapple with is how to make knowledge provision financially viable. Fee for service? That can become a retail operation, however, if service has to be customized to the needs of the client.

Add new comment