Existence of Informational Asymmetries - Essay Example

One of the very strong implications of the existence of informational asymmetries in the presence of heterogeneous product qualities in a market is the possibility of the better quality products being driven out by worse quality or unreliable products. This issue was first addressed by Akerlof (1970) who argued that if information was imperfect in that the buyers could not identify quality differences before the purchase was made, and sellers who were perfectly certain about the quality of the product being sold and had incentives to pass bad quality products as good ones, buyers would take the quality to be uncertain and accordingly base their decisions on the average quality of the products leading to a decline on the average willingness to pay. As a result the products which were actually of good quality would not receive their deserved prices and would thus be driven out of the market and the market would be filled up with the worse quality products or “lemons”. Akerlof demonstrated his point by using the example of the market of used cars. Suppose there are both good quality used cars as well as bad or defective ones, i.e, the lemons, on offer in the market and the consumer is not able to infer before hand which type is being offered to him by any particular seller. He therefore assumes it to be of average quality and is prepared to pay equal to the price of the average quality used cars. The owner of a good quality car therefore will not get the deserved high price for it as the customer has no way of knowing that it is in quality higher than the average. Therefore this good quality car shall be withdrawn from the market and so will be all other good quality cars, leading to a fall in the average quality of cars in the market. As a result of this downward revision of average quality, moderately good cars shall also fail to earn sufficiently high prices and will thus be withdrawn. Recurrence of this mechanism finally leads to a situation where only lemons are on offer in the market. To assess the validity of this proposition Bond (1982) attempted to empirically test it. In particular, the paper attempted to test the proposition that good products are driven out by bad products in the market for used pickup trucks. The amount of maintenance required on a particular truck was taken to be the measure of quality with the “lemons” being identified as used trucks that require significantly greater maintenance compared to the average. If the Akerlof argument is valid, then in this set up the result should be that only high maintenance trucks would be available in the market with the low maintenance ones being driven out due to the informational asymmetry arising out of the buyers’ inability to distinguish between high and low maintenance trucks beforehand, given the owners added information from experience1. Proceeding to test the proposition on these lines, the author found “if the effects of age and lifetime mileage on maintenance are controlled for, there is no difference in maintenance between trucks acquired new and trucks acquired used” (Bond, 1982). Therefore the “lemons” proposition appears to have been invalidated through this test. The possibility of counteracting institutional presence and relatively reduced informational gaps are forwarded as possible explanations to the violation the Akerlof proposition. The methodology of this paper however can be argued to have the following shortfalls: First, the test is done through simply comparing sample proportions which is not a very elaborate way of testing the lemons hypothesis. Secondly, while the data informs about whether any maintenance was undertaken in any of the following five categories of engine, transmission, brakes, rear axle, and other, the information on the exact costs of maintenance are not available in the data which do reduce the significance of the conclusions as the terms ‘high maintenance’ and ‘low maintenance’ lose meaning. Essentially, only the difference in frequency of repairs is considered leading to a failure in capturing the relative cost differences of maintenance. Moreover, this method of testing is not able to account for those trucks which have required major maintenance more than once within the interval of a single year (Pratt & Hoffer, 1984). A third difficulty of this method is that it requires controlling for the effects of age and total mileage on maintenance. Pratt & Hoffer, (1984) identifying these shortfalls adopt a modified methodology of testing the lemons proposition using the same data as Bonds. The modifications were in two dimensions. First, maintenance costs were directly estimated and second, trucks were grouped into classes by distinguishing between relatively new owners2 and longer term owners3. The Lemons proposition was tested through testing the hypothesis that the trucks that were purchased recently were lemons, that is, they required greater maintenance expenditure on average compared to the new and used trucks that the respective owners have held on to. The rationale for this hypothesis lies in the notion that owners who purchased their trucks new or used and have chosen to abstain from selling them have done so due to the existence of the lemons problem in that the sufficiently high prices would not be paid by the customers. The study found trucks transacted in a given time period to have average maintenance costs that were significantly higher compared to those not transacted. The difference in mean maintenance costs was found to be statistically significant and thus interpreted as evidence of the existence of the “lemons” problem in the given market. This according to the authors does deny the role of counteracting institutions as well as the possibility of a lessened information gap between the buyers and sellers. Therefore what emerges is that through a modified methodology of empirical testing, using the same set of data as Bond, Pratt and Hoffer (1984) arrived at a result that was directly opposite to the result arrived at by Bond (1982) and therefore supported the Akerlof proposition of Lemons. And this conclusion does seem to be more convincing particularly due to certain methodological advantages the study has over the Bond (1982) study. The Pratt and Hoffer methodology does seem to be adequately elaborate and more proper in case of testing the Lemons proposition. Apart from the shortfall of not computing the maintenance costs, it also requires controlling for age and total mileage and that has been done through excluding some particular vehicles from the test which itself can be argued to be restricting perfect and full modeling of the market. Not only does the Pratt and Hoffer methodology ably compute the average maintenance expenditures, it also scores above the Bond methodology by not requiring any sort of controlling for age, mileage or any other factors, the process of which hinders complete modeling. The most important critique against the Bond methodology is that it fails to capture the complete character of the lemons market. According to the basic proposition of the Lemons model the high quality products shall not be sold and thus comparing trucks acquired new to those acquired used shall be able to identify only those trucks which are “permanent lemons” while those with the “transitory lemon” feature shall not be identified. The Pratt and Hoffer system is able to capture both these elements of the lemons market. Therefore to conclude, it is pertinent to note that given the relative robustness of analysis and methodological advances, the conclusion of the Pratt & Hoffer (1984) article seems to be much more convincing and valid. References: Akerlof, George A., (1970). "The Market for Lemons: Quality Uncertainty and the Market Mechanism". Quarterly Journal of Economics 84 (3): 488–500. Bond, E.W., (1982) A Direct Test of the “Lemons” Model: The Market for Used Pickup Trucks, American Economic Review 72 (4): 836-840 Hoffer, G.E., and Pratt, M.D., (1984) Test of the lemons model: comment, American Economic Review 74 (4): 798-800 Read More