Wednesday, September 6, 2017

Perfect market in imperfect knowledge world?

Neoclassical economists assume decision making agents operate under perfect information. However imperfect information is the norm than exception of life (Hayek, 1945; Akerlof, 1998 and Stiglitz, 2001). Actually the problem is more serious than imperfect information, as the problem is imperfect knowledge. If human beings are rational beings, wrong decisions are only result of imperfect information or wrong information. However multipurpose rational mind is not what human posses and irrationality is the norm of life than exception (Cosmides and Tooby, 1994a, 1994b, 1996, 1987). Orthodox Economist think students fail in exam because they have bad teacher, who give them wrong information, but Conlisk (1996, page 672) stated

Though we teach that agents act as if unboundedly rational, we use gallons of red ink to inform students that they do not.

Lack of perfect rationality is well documented in wider literature of psychology, cognition, evolutionary psychology, institutional economics, experimental economics, behavioral economics and soon. See Cimoli, et al. (2006), Arrow (1969), Hodgson (2007); Nelson (2006), Williamson (1971, 1976, 1979, 1991, 1993, 1998, 2000, 2005), Hodgson (1998), Macher and Richman (2006), Smith (1991), Loomes (1998), Pagano (1999), Simon (1962, 1985), Vanberg (2002, 2004), Rutherford (1995), Cosmides and Tooby (1994a, 1994b, 1996, 1987), Nelson and Nelson (2002), Conlisk (1996), Rabin (1998), Camerer and Loewenstein (2004), Selten (1999), Goldberg (1990), Kahneman and Tversky (1982), Gigerenzer (1991), Clark (1918) Neale and Bazerman (1985), Cachon and Camerer (1996), Camerer et al. (1993), Akerlof and Yellen (1985), Aumann (1985), Grether (1990), Neelin et al. (1988) and Weg and Zwick (1991). 

Fortunately models with imperfect information may not be bad model for world with imperfect knowledge, when used with care. What knowledge enables us is to extract more conclusions from less information. The more knowledge we have, the more we can act from less information. The less knowledge we have the more information do we need to act on information (Mezgebo, 2014). Mind can be better understood as modular thinker or thinker which uses models or schemes (Piaget, 1936 and Pinker, 1997). The benefit of model is to infer more from less information at hand. As accuracy of the model declines the more information will be needed to act up on information. So knowledge problem is serious form of information problem and as result imperfect information models will be reliable proxy for knowledge problem as long as they did not underestimate level of information problem. However impact of imperfect information and impact of imperfect knowledge will have different channels of cause and effect. That is why it is better to use the imperfect information models with care.  Now let’s turn to the problem at hand.  

Imperfect information is observed, when people do not have perfect information about course of events. Asymmetric information is observed, when all concerned bodies are not having balanced share of information (Stiglitz, 2001).  Say a farmer may not know what the market price will be at harvest time and definitely his/her knowledge on weather condition could be very vague at best. As result s/he does have imperfect information or imperfect knowledge. When decisions are done under imperfect information or imperfect knowledge, it is hardly possible to assume they are the best decisions that can be made. The best example of asymmetric information is repair shops and health sector. The expert knows more than the customer and in this case you cannot assume the doctor and the repairman will never use his/her knowledge and information to cut better deal for him/her, while costing the society dearly. Under such reality it is hardly possible to accept market will result in best possible resource allocation to the society. The million dollar question is: does imperfect information of the real world make the conclusion of perfect information model of neoclassical useless fairytale[1]?           
 
One point raised by proponents of free market economy is that given people have to use some of their resource for information collection than for production, the level of output under risk will be lower than what can be under perfect market. But still market is constrained optimal. The point is that, if any organization can collect information, the market can do it better. In simple words nothing changes in conclusion of perfect market, except for welfare loss needed to avoid ignorance (Stigler, 1961). Stigler (1961, page 224) insisted 

Ignorance is like subzero weather: by a sufficient expenditure its effects upon people can be kept within tolerable or even comfortable bounds, but it would be wholly uneconomic entirely to eliminate all its effects. And just as an analysis of man’s shelter and apparel would be somewhat incomplete if cold weather is ignored, so also our understanding of economic life will be incomplete if we do not systematically take of account the cold winds of ignorance

The substance of the idea, as summarized by Stiglitz (2001), is that if we include the cost of information and accumulation of knowledge, perfect market is constrained optimal. The cost of the economy will increase by cost of information and accumulation of knowledge; but still market economy will provide the most possible higher welfare that can be generated from given resources. Imperfect market will not generate the best possible allocation of resources, which can be generated from perfect market only, but it will result in the next best allocation of resource. The point is that market is always efficient and this includes in dealing with problem of knowledge and information.

However this proposition is missing three important points. The first part of the problem is the fact that information is non rival good and market allocation of non rival good is highly unreliable (Stiglitz, 2001). The most important cost of information and also knowledge is producing it and the cost of sharing it is very negligible in magnitude. We should notice public goods and services are non rival and non excludable in character (Musgrave & Musgrave, 1980). So let’s take the two possibilities to share or not to share information. Either information becomes public good or not depends on how far it is shared. If information is shared it will create free riders as any public good (Stiglitz, 2001). Since market is inefficient supplier of public goods and services, as shown above, it will be self evident fact that market will not be optimal supplier of information.   

If information is kept hidden, there are two possibilities. One possibility is information could be under supplied in the market, because most people lack information which is owned by others. As result their decision will not be based on the best possible information and will not be the best decision possible. This could happen if the incentive for information collection does not encourage many people to collect information. We should notice the same information can be collected and supplied in cost effective manner by central body and as result market will not be optimal supplier of information. If incentives are very high to benefit from information and everyone is collecting private information, there could be over spending on information too (Stiglitz, 2001). The same information is collected again and again by different agents and this could not be the best possible information collection process. Small scale farmers may fail in first category of under informed agents and security traders in second category of over spending agents.  

Now let’s focus on second part of the problem. Information is convened not only by direct communication but also by action (Stiglitz, 2001). If information is kept private and other agents still can make it public by observing action, this will create free rider problem. When a big trader, who is known to invest on information, start buying grain, all small traders will follow his/her lead to benefit from his/her investment by being a free riders. So if s/he wants to make more profit, not only s/he has to hide the information but also s/he has to manipulate his/her behavior to give distorted signal to the competitors and free riders.

Far worst the well informed decision making agent does not have to hide the information or may not need to distort his action at all, but the agent should not be interested to share it. Given there is difference between knowledge and information, there could be still serious problem of understanding his/her knowledge. The small traders may understand his/her action but they may not be able to figure out what s/he knows. When everybody is dependent on everybody in such manner chaos is order of life and mass psychology will dictate the market dynamics to be chaotic and unstable. This will create distortion in the information and knowledge system itself. Market is not only inefficient supplier of information but is serious source of information distortion and mass absurdity.

Let me use simple example to explain it. Think about two cars, which are travelling in opposite sides of one road. If their drivers have perfect information about the intention of other driver, one will take left and the other to avoid collusion will take right. The self interest of each driver, to avoid collusion, will result in efficient use of the road. However the perfect information scenario assumed above is unrealistic. Normally drivers will have highly imperfect information and have to evaluate the intent of other driver from his/her action. If everyone is not promoting his/her own welfare, or is not trying to move faster in the road, one can stop and wait for the first car to pass and choice his/her path then after. This will reduce the efficiency of transportation, but will lead to low collusion. When people are promoting their own welfare, free competition between drivers to use the road will end in collusion, in most cases. Collusion free driving under such reality is an exception than norm. The problem is that not only each driver does not know the intention of other driver, but also action and reaction of each driver will create more confusion in the system. Have you ever try to pass over narrow path when another person is coming from the other side. Imagine how chaotic the system becomes! Competition to use the road will crate chaos of misunderstanding. Every good intention will be lost in translation to make collusion highly likely. This is why promotion of self interest under imperfect information will not only result in Pareto sup-optimal resource allocation, but it will also create more imperfection of information and knowledge in the system.  

This is why market economy is not only facing imperfect information, but is also creating additional imperfection of information. Distortion can also come from wrong presentation of contract. Say worker will over sell him/her self, firms will over estimate the benefit and under estimate the cost of their goods and services and soon.

The third source of the problem is much serious and dangerous. The market process which promotes selfishness is catalyst of such moral decay. After all a doctor may do what is best for patient, a mechanic may do what is best for the car owner, a worker may speak honestly and firms may be ethical enough to be frank and trustworthy. When all life becomes about competition and ‘virtue of selfishness[2]’, the market will be catalyst for distortion of information and knowledge. In this reality, can we say market competition will give us ideal reality and the best possible organization possible? Definitely no! This is why information economists like Stiglitz (2001, page 7 and 12) stress the fact that

Perfect competition - model was not robust – even slight departures from the underlying assumption of perfect information had large consequences…. In the standard paradigm, the competitive general equilibrium model, there were no shocks, no unanticipated events: at the beginning of time, the full equilibrium was solved, and everything from then on was an unfolding over time of what had been planned in each of the contingencies.
  
So it is right to conclude that “Under the imperfect information paradigm, markets are almost never Pareto efficient” (Stiglitz, 2001, page 15). However we should remember let alone perfect market and especially the virtuous market, even devil need to have a devil advocate. The above analysis is not complete because there is learning happening over time and there is collective rationality which often can over perform individual rationality and sometimes destroy humanity under wrong mass psychology. In simple words the above analysis ignores one important fact that signaling can be also supplied from right institutions.

The idea is people learn from their action and will refine facts from fiction. I may hire wrong worker but who has best interest to get it right? The answer is me, not government or other institution. As long as there is correct signaling of mistakes, market can do it better. For example in experimental market, agents are given fragmented information of the whole system. In this case all the information is available in the economic space, but everyone has small portion of the complete information. Then they are allowed to buy and sell so they can make profit. In dynamic manner market price will summarize their information to give them right signal. If you think price should be high but it turns out to be low, you learn from it. In dynamic manner the market will take you toward rationality or outcome that would be expected by well informed central planer. Such simple markets are often observed to attain not perfection but optimality in terms of discovering the right market price (Plott, 1981, 1986; Plott and Sunder, 1988; Smith and Williams, 1981; Roberson and Smith, 1982; Smith, 1962, 1964, 1967, 1982, 1990, 1991; Easley and Ledyard, 1983; Conlisk, 1983; Bray, 1982; Knez et al., 1985; Friedman, 1984 and Gode and Sunder, 1993). The point of those experiments is that when right signaling institutions like double auction are used and when agents do not have to be burdened with predicting the action and intention of other agents, market will create market rationality using bounded rationality of the decision making agents.      

We should notice those experiments did prove the need to have societal view of rationality, beyond individual view of rationality which dominates the dominant schools of psychology. After all we do learn from mistakes as long as there are corrective institutions at work. Natural science is science not because Newton is Angel, or Einstein is perfectly rational, it is science because the scientific system and culture works in long run. The rationality of science does not emanate from human rationality, it emanates from the system which organizes the thinking and experimenting agents. The strength of evolution does not emanate from far sight rationality of genes or their phenotype but from its cumulative selection for fitness that lead to path dependent accumulation of fitness. The same fitness selection is observed to work in market with right kind of institution. However those experiments did not really replicate the problem as we see it.

First they ignore the fact that information may not be collected efficiently. Their focus was on how market can assemble information from different agents by using profit. However they assume all information is in the system. If I know more than the market, I will bet against the market or other decision making agents. In making profit I will supply the information to the market. The market does not know my information in detail, but it will incorporate my information in its decision making process. Say I know there is more demand than supply, as result market price has to be higher. If market price is kept low, I will buy and store the commodity. This will push price up because I increase demand. This process will stop when my information is reflected in the price of the market. While I am rewarded for my foresight, other agents who used to follow (influence) the market dominantly will learn their mistake by facing loss[3]. However this will not solve all problems of imperfect information and imperfect knowledge. 

In real world information has to be collected and the question was do people have incentive to collect optimal amount of information in least cost possible? If there is less than optimal and distorted information in the system, there is no grantee the market will generate efficient price discovery[4]. Actually the same experimental studies cited above did observe the fact that when agents have to predict other people’s behavior, the market system will have bubbles and busts and very low level of efficiency. Far worst the experiments cannot show and did not show that every decision made by the subjects of the experiment is rational and optimal, as assumed in neoclassical text books. Future experiments need to study this dynamics from better angle to give us better picture which can replace the endless bickering of Greek mythology which dominates most economics.

The experimental studies did however show institutional design matter to attain optimality (Plott, 1986; Smith, 1964, 1967, 1982, 1990; Gode and Sunder, 1993; Roberson and Smith, 1982 and Easley and Ledyard, 1983). Under wrong signaling by wrong institution you can send people to hell or under right signaling by right institutions you can send them to heaven. In simple words, institutional design does matter for collective optimality. The question therefore is what kind of institutional design can make otherwise perfect market with imperfect information constrained Pareto optimal.

The answer we have until this point includes complete and efficient credit, information, insurance and future markets (Stiglitz, 1981). To understand this let’s make it simple to understand from practical point of view. Farmers have to choice which input to use and which output to produce, without knowing how much output they can get from each effort. Different inputs have different productivity under different weather conditions, which are behind the control of the farmer. For example, fertilizer can give different level of output based on moisture content of the soil, which depends on natural weather conditions. At the same time differ grains has different resistance to different shocks. That is why there is high output uncertainty in farming. Unfortunately, given necessity nature of most agricultural products which lead to low price elasticity of demand, random change in output will generate more than proportional change in price[5]. Means production decisions have to be done in face of high output and price risk. If market forces can supply optimal information to decision making agents, market based allocation of resources will be, still, constrained optimal decision. But given the fact that information is public good, market forces has tendency to under supply it. As result, farmers have to make production decision with less than optimal information. This is why optimal and cost effective information has to be supplied by information markets, which does not have to be competitive market (Stigler, 1961). The efficient market hypothesis of Read (2013) implies that information can be assembled by market forces efficiently and by implication it lead to conclusion that collecting information does not pay. However from where does the market get the information if information does not pay at all is unanswered question? It is another badly cooked Greek mythology of economists. The idea do give good presentation of the good information processing capacity of market but to say it is efficient is to hide beyond lie not to see the bubbles and busts of market and market absurdity which is daily news of this world. Ones we separate information from knowledge, in addition to incentive problem of information collection, the theory will become another badly under cooked bread which tests ugly. The bad cook, which is Read (2013), did destroy the nutrition which is definitely inside the grain that is supposed to be bread.

Even if there is enough information in the market, market allocation needs additional institutions to be constrained Pareto optimal. First the information has to be squeezed in to simple values that can be easily understood by decision making agents. This is where the nutrition of Read (2013) becomes useful. Spot information will be summarized to spot prices by spot market as explained above. However future information has to be analyzed efficiently to make the best prediction possible. Future markets are needed to squeeze all information in the economy in to simple to understand future prices (Mishkin, 2004).

If Teff future of one quintal with strike date of 3 month is sold at strike price of 1000 birr. It means buyer and seller are agreeing to exchange 1 quintal Teff after 3 months for 1000 birr. If someone has information that shows that Teff spot price is going to be 900 after 3 month, s/he has to sell future at strike price of 1000 birr. After 3 months s/he will buy the grain in spot market for 900 birr and will honor its future contract at 1000 birr. As result s/he will make a profit of 100 birr. When most people, with information that spot price is going to be 900 birr after 3 months, start future selling, future price will decline to ward 900 birr. Viewed from different angle, if a person believes that spot price after 3 months is going to be 1100, s/he will buy in future market for strike price of 1000 birr. After 3 months s/he will pay 1000 birr for one quintal Teff based on its future contract and will sell it at 1100 birr in spot market, to earn a profit of 100 birr. At current time when many people with such expectation, start buying to the future, future price will increase to ward 1100 birr. One way or another decision making agents have incentive to squeeze whatever information they have in to future prices. This is why future trading is needed to make sure that decisions are done based on all possible information and all information is squeezed into few prices that decision making agents can understand. This will benefit not only those which participate on such activity but also those who watch future prices to get the best prediction possible about future prices. When there is efficient and well functioning future market with adequate information it will give us the best prediction possible of future prices. A farmer than using his/her limited information to make production decision for grain that will be sold after months at who know the price, they can use the best prediction possible (Mishkin, 2004). If decision is done in the best information possible what can be changed then after (Read, 2013)?      

Read’s (2013) efficient market hypothesis is wrong in many grounds. First we can understand part of his problem from type I and type II errors of decisions. Science is conservative and it often rejects good ideas in trying to protect itself from countless wrong ideas. Science often faces high type I error as result. Highly competitive market is highly progressive and will try to milk a lot of information to make best prediction. As result type I error often is not serious problem. But it reacts to every junk of information as result it is very unstable system. It faces serious type II errors. There is too much unnecessary information and imperfect knowledge. In addition when knowledge is incomplete and uncertainty very high, information and knowledge of your follow human beings and the market itself will become useful guide under mass psychology and market reinforcement. Not because market or follow traders are right all the time, this is because decisions are done at market level at competitive system and the entire market takes them seriously in carrot and stick process that follows. Investors may run away and banks may reject to lend when the market valuation turn to be bad. The market as its dominance rises, it has tendency to fulfill its own prophecy but it does not grantee market is always right. This is why bubble and bust are daily reality of competitive markets and it is highly likely good firms often are forced to exist market because of security market failures which wrongly rate them low. Keynes after all is quoted in saying (Wikiquote, 2017)[6]

The ignorance of even the best-informed investor about the more remote future is much greater than his knowledge, and he cannot but be influenced to a degree which would seem wildly disproportionate to anyone who really knew the future, and be forced to seek a clue mainly here to trends further ahead. But if this is true of the best-informed, the vast majority of those who are concerned with the buying and selling of securities know almost nothing whatever about what they are doing. They do not possess even the rudiments of what is required for a valid judgement, and are the prey of hopes and fears easily aroused by transient events and as easily dispelled.

However the problem does not end there. In one side market with functional and efficient information market will never supply perfect information. Some information is not perfectly available like weather condition and other information is too costly to collect compared to the benefit it can generate. In other side future markets may not be highly efficient under such reality of imperfect information. Experimental studies using future markets did show the fact that the market has tendency to create bubbles and booms or cyclical instability (see Smith, 1962, 1964, 1967, 1982, 1990, 1991). This can be seen from the fact that future prices are highly unstable and will create serious bases risk as result (Mishkin, 2004). Basis risk is difference between future spot price and the prediction of future trading about it in form of future price. If teff price is predicted by future market to be 1000 after 3 months, and after 3 months the spot market price turn out to be 1300 birr, those who sell to the future will loss 300 and future buyers will gain 300 birr. Basis risk is risk of future market failing to predict future spot price. Future markets in practice do reduce spot risk at cost of increased bases risk. If all information is not used and if the system also creates instability, there is risk that has to be managed efficiently.  

That is why insurance markets must be functional to allocate risk from risk averse individuals toward risk takers, for price. If a farmer can produce a Teff valued at 10, 000 birr from input cost of 8, 000 birr, there will be a profit of 2, 000 birr. However if the farmer is risk averse and needs a risk premium of 3, 000 birr to assume the risk on his/her 8, 000 birr, s/he will not engage in the  productive activity. As result value of output will decline by 10, 000 birr. Now let’s say there is someone who can give the farmer full insurance of 10, 000 birr for premium of 1, 000 birr. So to achieve the best possible allocation of resource, the farmer needs to buy insurance for 1, 000 birr to produce and earn net profit of 1, 000 birr. This could be functional option trading or some sort of insurance scheme. Option trading gives you right to exercise a future contract when it is favorable to you without duty when it is not favorable to do so (Mishkin, 2004). You have no basis risk of participating in future market if you buy option, except you have to pay risk premium in buying option. The option seller (writer) is betting on the assumption you will not exercise the option and s/he will have your payment of risk premium as income as risk taker. This is one way to efficiently trade risk between risk averse and risk taker individuals. Production risk and other risks can be also handled by efficient insurance schemes. That is why for market allocation to be constrained Pareto optimal under risk, there must be functional insurance market or similar institution. We should understand the markets that we have in form of future and option trading are functional but far from Pareto optimal (see Mishkin, 2004). As result the conditions stated for existence of efficient market hypothesis cannot be satisfied in real world. 

Finally, there must be function finical intermediaries or banks, too. People must able to stabilize their investment and consumption in face of stochastic nature of their income, spending and wealth. In general even if there is risk in a market, but there are

1.      adequate information in the market from highly information market
2.      Highly functional future market
3.      Highly functional insurance market or related institutions  
4.      Highly functional intermediaries (banks)

Market allocation will be the next best possible (constrained Pareto optimal) resource allocation and there is no need to change things to improve it farther. As functionality declines market allocation do need additional institutions to make it work better. Say telling doctors serving people is our duty, so to frame their mind, is needed at least. Under complete set of highly functional markets which are preferably perfect, decisions will be done based on the best possible information and all risks will be efficiently managed by financial markets (insurance and banks). Under such reality there is nothing that the state or other institutions can do to improve farmers’ decision than to respect market allocation.

The challenge is the same problems which make grain markets inefficient will in turn reduce efficiency of above institutions to make the entire system Pareto suboptimal and sometimes dysfunctional (Mishkin, 2004). Those problems are the primo facto for those who cry let the state free and this is where they start telling their Greek mythology that explain how self scarifying Santa known as functional state give fire to humanity by stealing it from the Gods of Olympus and as result is seriously hunted by the Zeus of our world who want to kick the ladder of development from rest of the oppressed people.    

Akerlof (1998), for example, insist that under asymmetric information missing markets are to be expected. If there is asymmetric information on quality of products that can lead to adverse selection problem the market may fail to exist despite the fact that demand and supply coexist and there is price range which is beneficial for both buyer and seller. This is because since adverse selection will increase the risk to above average risk, high premium will be asked for that, but this will push the less risky farther out of the market pushing the risk up and the premium up and finally this vicious circle will eliminate the market. How spot market, future market, option market, insurance market, financial intermediaries and other institutions under competitive system will not be perfectly efficient can be explained by the same logic of imperfect and asymmetric information used to explain their genesis above (see Mishkin, 2004 and Stiglitz, 2001 for review).    

This is why under imperfect and asymmetric information to expect fully functional markets, let alone perfect market or extremely efficient market, is highly unpractical. Imperfect information and asymmetric information which leads to moral hazard and adverse selection problem is the reason why most financial institutions come to picture. However imperfect information and asymmetric information will also reduce their capacity to solve all financial problems too. In general market forces by themselves will not solve the information problem and alternative institutions including state has to come to picture to find solution for problems faced by humanity. If there is location where the invisible hand, with its full set of functional institutions, becomes useless fairytale, this location is to be found in under developed rural areas of developing economies.    

This is why market, as institution, is very important, but not perfect institution to serve as perfect recipe for every developmental problem. Even it is surprising to notice that humanity in 21st century is debating if market makes mistake or not? One thing common among philosophers, social scientists and practitioners is that in order to show what they believe is perfect, normally will end up proving what they believe is wrong. This is unfortunate reality that we have, where the practical good becomes the archenemy of none existing   perfection. In market economics in order to show that the market is perfect, they make it easy for others to prove that the market is imperfect and to demand for its replacement.

There is no doubt that market is the most effective institution in allocation of resource, compared to any alternative institution in most cases. However it is tool which has to be used when it serves us and which has to be managed, perfected and discarded when it is our interest to do so. Market like state is our servant not our master, unless we start worshiping them like the intellectual slaves of the mad king or fire stealing ladder sharing  mythical heroes of the let state free moment. All is just Greek mythology, but you have to collect the nutrient inside and cook it better.


[1] It is important to note this part is highly dependent on Stiglitz (2001)
[2] Read Rand (1964) for unapologetic defense for selfishness.
[3] The same result of improved rationality through computer learning is documented by Clark and Karmiloff-Smith (1993) too. However contradictory evidence is given by Rabin (1998) in stating economist did not care about irrationality assuming learning will improve out come and in markets were experts with adequate experience will act as rational agent. How experiments which checks weather knowledge of statistics improves decision making or not does not show any support for experts’ better judgment or the idea of the that people learn in the real world. Moreover studies show that people do still make mistakes in environment that they are familiar. Experts are over confidence compared to layman when they make right or wrong prediction. When their information is not adequate to make any prediction, experts have tendency to make predictions and to be very confident about it.

However Rabin (1998) seems to be stack in his classic school of psychology rejecting the real world as reality smiles on his foolishness. Is over confidence itself out of check and balance? To say people are not perfectly rational forever is not the same as to say they will never learn at all. But Rabin (1998) himself did notice people who are expert on their field of study do make mistakes, but it is absurd to think Rabin will never learn forever. As long as there is competition such over confident will be also refined cannot be rejected to justify the learning process having tendency to move toward rationality (Aumann, 1985). But those results do have critic too. For example Camerer and Loewenstein (2004) pointed an idea that since repetitive transactions are not common in real life, experimental outcomes are not applicable to real life. However the solution is we should devise institutions like double auction markets to make them repetitive and efficient. Game theory analysis by Blume (1982), Plott and Sunder (1988) and actual market analysis of Timmerman (1994) also found that although learning improves performance there are multiple optimal solutions and not necessarily rational expectation equilibrium. But we are not expecting perfection like neoclassical economics but only tendency to move in the right direction.
[4] For extreme and unpractical defense of market efficiency under efficient market hypothesis see Read (2013).  
[5] Stiglitz (1981) point out that as long as price elasticity is between 0.5 and 1, price instability will be coupled with income stability. Means price variability will stabilize income and farmers will not face high risk due to production and price risk. What it means is either farmers produce more or less their income will be stable around average. The problem is farmers will not have incentive to produce more because it does not pay to do so. To imply production and marketing risk does not matter under such reality is absurd.

No comments:

Post a Comment

Culture and Development

While most economists remain locked in a historical tug-of-war—cycling between the worship of Zeus (the market) and Prometheus (the state)...