柠檬市场：质量的不确定性和市场机制(中英对照

1、The Markets for Lem ons ”Quality un certa inty and The Market Mecha nism柠檬市场：质量的不确定性和市场机制Geogre A. Akerlof阿克洛夫一、引言This paper relates quality and un certa in ty. The existe nee of goodsof many grades poses in terest ing and importa nt problems for the theoryof markets.（本文论述的是质量和不确定性问题。现实中存在大量多种档次的 物品

2、给市场理论提出了饶有趣味而十分重大的难题）On the one hand, theinteractionof quality differences and uncertaintymay explain importantin stitutio ns of the labor market.（一方面，质量差异和不确定性的相互作用可以解释劳动力的重要机制）On the other hand, this paper prese nts astruggli ng attempt to give structure to the stateme nt: Bus in ess in under-deve

3、loped countries is difficult; in particular, a structure isgive n for determ ining the econo mic costs of dish on esty.（另一方面，本文试图通过讨论获得这样的结论：在不发达国家，商业交易是困难的，其中，特 别论及了欺骗性交易的经济成本）Additio nal applicatio ns of the theoryinclude comments on the structure of money markets, on the notion ofinsurability, on

4、 the liquidity of durables, and on brand-name goods.（本文的理论还可以用来研究货币市场、保险可行性、耐用品的流动性和名牌商 品等问题）There are many markets in which buyers use some market statistic tojudge the quality of prospective purchases. （在许多市场中，买者禾寸用市场的统计数据来判断他们将要购买的商品的质量）In this case there isincen tive for sellers to market poor q

5、uality mercha ndise, since thereturns for good quality accrue mainly to the entire group whose statistic is affected rather tha n to the in dividual seller. As a result there tends to be a reducti on in the average quality of goods and also in the sizeof the market.（在这种情况下，卖者有动力提供低质量商品，因为某种商品的 价格主要取

6、决于所有同类商品质量的统计数据而非该商品的实际质量。结果，商品的平均质量将趋于下降，市场规模将不断缩小）It should also be perceivedthat in these markets social and private retur ns differ, and therefore,in somecases, gover nmen tal in terve nti onmay in crease the welfare of allparties. Or private institutions may arise to take advantage of thepote n

7、tial in creases in welfare which can accrue to all parties.（我们还可以观察到在这种市场上，对个人和社会有不同的回报，因此，某种情况下， 政府的干预可以增进社会的总体福利水平。或者说，私人组织利用了整个社会福利水平潜在的增长机会而使自己获利）By nature, however, these institutionsare non atomistic, and therefore concen tratio ns of power- with illcon seque nces of their own-can develop.（实际上

8、，这些私人组织的行动所产生的影响并非可以忽略不计，因此，尽管集权本身有许多负面影响，但是，一定 程度上的集权可以保证经济的健康发展）The automobile market is used as a fin ger exercise to illustrate and develop these thoughts. It should be emphasized that this market is chosen for its concreteness and ease in understanding rather than for itsimportanee or realism.（

9、下面，我们用汽车市场作为例子来阐释和进一步发展 以上的思想。需要指出的是，之所以选择旧车市场来讨论，是因为这个例子很具 体，并且容易理解，而不是因为它的重要性和有何现实意义）二、以汽车市场模型为例（一）汽车市场The example of used cars captures the esse nee of the problem. Fromtime to time one hears either mention of or surprise at the large pricediffereneebetween new cars and those which have just left

10、 the showroom.The usual lunch table justification for this phenomenon is the pure joy of owning a new car. （旧车市场的例子可以抓住问题的本质。人们不止一次 的听说或惊讶于新车与刚刚开出样品陈列室的汽车之间的价格存在巨大差别。 对 于这种情况最普遍的解释是人们有一种对“新”车的特别偏好） We offer a different explanation. Suppose （for the sake of clarity rather than reality） that there are

11、 just four kinds of cars. There are new cars and used cars. There are good cars and bad cars （which in America are known as lemons）. A new car may be a good car or a lemon, and of course the same is true of used cars. （假设（作出这种假设只是为了简化分析，而非从 实际出发）用四种汽车：新车和旧车；高质量的车和低质量的车（低质量的车在 美国被称为“柠檬”）。一辆新车可能是高质量的，

12、也可能是“柠檬” ，当然一辆 旧车也同样有两种情况）The individuals in this market buy a new automobile without knowing whether the car they buy will be good or a lemon. But they do know that with probability q it is a good car and with probability （1-q） it is a lemon; by assumption, q is the proportion of good cars produced

13、 and （1 - q） is the proportion of lemons. （在这样的市场上，消费者买新车时并不 知道车辆到底是高质量的还是“柠檬” ，但是他知道这辆车是高质量的概率是 q， 是“柠檬”的概率是“ 1-q ”。这里假设q是高质量车占所有汽车的比率，1-q是“柠檬”的比率）After owning a specific car, however, for a length of time, the car owner can form a good idea of the quality of this machine; ., the owner assigns a ne

14、w probability to the event that his car is a lemon. This estimate is more accurate than the original estimate.（车主在拥有汽车后一段时间内就会了解到该车的质量， 也就是说， 此时车主可以赋予该车可能是 “柠檬”的一个概率值，这个估计比初始的估计更加准确） An asymmetry inavailable information has developed: for the sellers now have morekno wledge about the quality of a ca

15、r tha n the buyers. But good cars and bad cars must still sell at the same price- since it is impossible for a buyer to tell the differe nee betwee n a good car and a bad car.（于是，信息不对称发生了：卖主掌握了比买主更多的关于汽车质量的信息。对于买主来 说，由于他不能在买车时区分汽车的质量，所以，高质量车和低质量车只能以同一个价格水平出售）It is appare nt that a used car can-not h

16、ave the samevaluation as a new car - if it did have the samevaluation,it would clearlybe advantageous to trade a lemon at the price of new car, and buy another new car, at a higher prob-ability q of being good and a lower probability of being bad.（显然，一辆新车和一辆旧车不应该有相同的评价。如果他们拥 有相同的评价，车主通过以高质量车的价格出售“柠檬

17、”后再买一辆新车，该新 车是高质量的概率q大于原先购买新车的概率）Thus the own er of a good machine must be locked in. Not only is it true that he cannot receive the true value of his car, but he cannot eve n obta in the expected value of a new car.（这样，高质量车的拥有者将不会卖出汽车，因为如果卖出汽车他不 仅无法得到汽车的真实价值，也无法获得一辆新车的期望价值）Greshams law has madea mo

18、dified reappearanee. For most cars traded will be the lem on s, and good cars may not be traded at all. The bad cars tend to drive out the good （in muchthe sameway that bad moneydrives out the good）.（这里出现了一个修正后的格莱欣定律。当“柠檬”充斥汽车市 场时，高质量车根本不会成交，于是，如同劣币驱逐良币一样，低质量车将高质 量车逐出市场） But the an alogy with Gresha

19、ms law is not quite complete: bad cars drive out the good because they sell at the same price as good cars; similarly, bad money drives out good because the excha nge rate is even.（然而，该结果与格莱欣定律并不完全一样：低质量车驱逐高质量车是因 为低质量车和高质量车在相同的价格水平上出售。相似地，劣币驱逐良币是因为两者的交换禾率相同）But the bad cars sell at the same price as

20、 good cars since it is impossible for a buyer to tell the differenee between a goodand a bad car; only the seller knows. In Greshams law, however, presumably both buyer and seller can tell the differeneebetween good and bad money.So the an alogy is in structive, but not complete.(但是低质量车之所以与高质量以同一个价格

21、出售是因为买主无法区分一辆车到底是低质量的还是高质 量的，这只有卖主才知道的信息。在格莱欣定律中，买主和卖主大概否可以区分 劣币和良币。所以说，这个类比具有启发性，但非完全相同)(二) 非对称信息It has bee n see n that the good cars may be drive n out of the marketby the lem ons. But in a more con ti nu ous case with differe nt grades ofgoods, even worse pathologies can exist. (我们已经看至卩质量好的汽车会被“

22、柠檬”逐出市场。但是，在不同质量等级的产品连续分布的情形下，更糟糕 的情况也会发生) For it is quite possible to have the bad driving out the not-so-bad driv ing out the medium driv ing out the no t-so-good driving out the good in such a seque nee of eve nts that no market exists at all.(质量最差的汽车在将质量最好的汽车逐出市场后，会继续将质量较好、质量中等和质量稍差的汽车依次逐出市场，从而导

23、致二手车交易市场根本无法存在)One can assume that the dema nd for used automobiles depe nds moststrongly upon two variables - the price of the automobile p and the average quality of used cars traded,卩,or Q d = D (p,卩).(假设对二手车的需求主要取决于两个变量二手车的价格p和进入交易的二手车的平均质量卩，即 Cf = D (p, 卩)Both the supply of used cars and also t

24、he average quality 卩 will depend upon the price, or卩=卩(p) and S=S(p). (而二手车的供给和平均质量卩都取决于二手车的价格，即卩= u(p)和S=S(p) And inequilibrium the supply must equal the demandfor the given average quality, or S(p) = D (p, 卩(p). As the price falls,normally the quality will also fall. And it is quite possible that

25、no goods will be traded at any pricelevel.(在均衡状态下对于给定的平均质量，二手车的需求等于供给，即S(p)= D (p, u(p)。随着二手车交易价格下降，进入交易的二手车的平均质量当然也 随之降低，最终导致在任何价格水平下都不存在二手车交易)Such an example can be derived from utility theory. Assumethat there are just two groups of traders: groups one and two. Give group one a utility function :

26、(效用理论可以用开帮助我们解释二手车市场的例子。假定市场上有 两类交易者，交易者1和交易者2,交易者1的效用函数为)a山八1 + Jxii - 1WhereMis the consumption of goods other than automobiles, I： I is thequality of the i th automobile, and n is the nu mber of automobiles.（其中，M代表交易者1对二手车之外其他商品的消费， Ad證一阴汽节的质量，n 为汽车的数量）Similarly, let （同理，交易者2的效用函数是）II也二M +鸥kZ iTh

27、ree comme nts should be made about these utility fun cti ons:（对于这些效用函数有三点需要说明）（1） without lin ear utility （say withlogarithmic utility） one gets needlessly mired in algebraic complication（第一，这里采用线性效用函数（如对数效用函数）来讨论问题，这样我们就无需处理复杂的代数运算）（2） The use of linear utility allows a focus on the effects of asym

28、metry of in formati on; with a con cave utility fun ctio n we would have to deal joi ntly with the usual risk-varia nee effects ofun certa inty and the special effects we wish to discuss here.（第二，现 行效用函数还能让我们专注于对非对称信息的效果。如果在凹形效用函数下进行研究，我们将不得不在考察我们最希望关注的事实的同时，考虑不确定性所带来的风险变动的影响）（3） U i and U 2 have th

29、e odd characteristic that the additi on of a sec ond car, or in deed a kth car, adds the same amount of utility as the first.Aga in realism is sacrificed to avoid a divers ionfrom the proper focus.（第三，U和12具有边际效用不变的特征，即增加第二 辆车的消费所带来的效用的增加都与增加第一辆车消费所带来的效用增加相等。 在此，我们再一次为了研究的需要牺牲了边际效用变动的现实）To con ti nu

30、e, it is assumed （1） that both type one traders and typetwo traders are Von Neuma nn-Morge nstern maximizers of expected utility;（接下来假设：（1）两类消费者都是冯诺依曼一摩根斯坦效用函数中的追求期望效用最大化者） （2） that group one has N cars with uniformly distributed quality , U 工:-,and groupt wo has no cars;（2）交易者 1 拥有 N辆质量为b并且服从均匀分布的汽

31、车，其中，交易者2没有汽车）（3） that theprice of other goods M is unity.（3）把其他商品 M的价格看成为一个单位）Denote the in come （in clud ing that derived from the sale of automobiles） of all type one traders as Y 1 and the in come of all type two traders as 丫2.（把各种类型的交易者1的收入（包括销售二手汽车所带来的收入）记作 Y, 将交易者 2 的收入记为 丫2）The dema nd for u

32、sed cars will be the sum of the dema ndsby both groups. Wheno ne ignores in divisibilities,the dema ndforautomobiles by type one traders will be（对二手车的需求即是两类交易者对二手车需求的总和。当我们忽略不可整除的可能时，交易者1对二手车需求的 数量可以表示为）Di 二 丫1巾口 /p1Di = 0l/p 1（交易者1对And the supply of cars offered by type one traders is二手车的供给是）S2 = p

33、N/2p总2With average quality（二手车的平均质量为）m 二 P/2（To derive （1） and （2）, the uniform distribution of automobile qualityis used.）（在（1）和（2）两个表达式中，我们使用了二手车质量服从均匀分布这一假设）Similarly the dema nd of type two traders is（同样，交易者 2 的需求为）|D2 = 03m/2 p|And （供给是）陽=0Thus total dema nd D （p, u） is（因此，总需求 D （p, u）是）|D（山 u）

34、 =（丫总十 Vj/P如果D uD（p, u）=Yi?/p如果 u p However, with price p, average quality is p/2 and therefore at no price will any trade take place at all: in spite of the fact that at any give n price betwee n 0 and 3 there are traders of type one who are willi ng to sell their automobiles at a price which trade

35、rs of type two are willi ngto pay.（尽管有这样的事实，即在0到3之间的某个价格下有的交易者1愿意 卖掉他们的车，而有的交易者2愿意在这一价格下购买。但无论如何，在价格为 P，平均质量为p/2这种质量均匀分布的情形中，在任何价格水平下都根本不会 发生二手车交易）(三) 对称信息The forego ing is con trasted with the case of symmetric in formati on.Suppose that the quality of all cars is uniformlydistributed, 0 兰 x w Then

36、 the dema nd curves and supply curves canbe written as follows:(上面的分析将与对称信息下的情形形成对照。假设所有汽车的质量是均匀分布的，0至蔑兰2。那么，供给曲线和需求曲线就可以 写成如下形式：)Supply (供给曲线是)S(p)=Np1S(p)=0p1And the dema nd curves are(需求曲线是)D(p)=(Y1+Y2)/pp1D(p)=(Y2/p)1p3/2In equilibrium(均衡时)P=1如果 Y2N(3)P=Y/N如果 2Y/3NvY (4)p=3/2如果 N2Y/3(5)If N Y, i

37、n which case the in come of type two tradersis insufficient to buy all N automobiles, there is a gain in utility ofY2/2 units.)(如果N Y,此时交易者2的收入不足以购买到N辆汽车，那么，将有 Y/2 单位的效用收益)Fin ally, it should be men ti oned that in this example, if traders of groups one and two have the sameprobabilistic estimates a

38、bout the quality of in dividual automobiles- though these estimates may vary fromautomobile to automobile- （3）, （4）, and （5） will still describeequilibrium with one slight change: p will then represent the expected price of one quality un it.（最后，对于该例子应该提到的是，如果两类交易者对每辆汽车的质量具有相同的概率估计一一尽管这种估计因汽车的不同而不 同

39、一一（3）、（4）和（5）仍将能描述均衡，只不过此时p将代表每单位质量的期望价格）三、例证和应用（一）保险It is a well-known fact that people over 65 have great difficulty in buying medical in sura nee. The n atural questi on arises: why does nt the price rise to match the risk?（总所周知，年龄超过 65岁的人很难买到医疗保险。这就产生了一个问题：为什么保险费不能一直上升到风险相当的水平？）Our an swer is th

40、at as the price level rises the people who insurethemselves will be those who are increasinglycertain that they will needthe in sura nee; for error in medical check-ups, doctors sympathy with older patients,and so on makeit mucheasier for the applicant to assessthe risks invo Ived tha n the in sura

41、nee company. （我们的回答是，当保险费提高时，只有那些越来越确信自己有必要投保的人才会买保险。医疗体检中的弊端、医生对病人的同情等因素，使得保险申请人比保险公司更加了解自己的风 险状况）The result is that the average medical condition of insuraneeapplicants deteriorates as the price level rises -with the result that no in sura nee sales may take place at any price. （结果，当保险费上升时，投保申请人的平

42、均健康状况不断恶化，导致最终将不会在任何保险费水平上签订 保险合同）This is strictly analogous to our automobiles case, where the average quality of used cars supplied fell with a corresp onding fall inthe price level. This agrees with the explanation in insurance textbooks: （这与我们汽车市场的例子十分相似， 在那里，待售旧车的平均质量随着市场价 格的下降而下降。关于这一点，保险学教科书的

43、解释是这样： ）Generally speaking policies are not available at ages materially greater than sixty-five The term premiums are too high for any but the most pessimistic （whichis to say the least healthy） insureds to find attractive. Thus there is a severe problem of adverse selection at these ages.（通常， 65 岁

44、以上人的保单是无效的。此时，保险费太高以至于除了那些最悲观的 （即健康状况最差的） 投保人而没有其他人会对投 保感兴趣，因而在这一年龄存在着严重的逆向选择问题）The statistics do not contradict this conclusion. While demands for health insurance rise with age, a 1956 national sample survey of 2,809 families with 8,898 persons shows that hospital insurance coverage drops from 63

45、per cent of those aged 45 to 54, to 31 per cent for those over 65.（以上结论与统计数据是一致的。 1956年对 2809个家庭 8898个人进行的全 国性调查问卷显示， 当人们随着年龄的增长越来越需要保险时， 医疗保险的覆盖 率从 4565 岁人群的 63%下降到 65岁以上人群的 31%） And surprisingly, this survey also finds average medical expenses for males aged 55 to 64 of $88, while males over 65 pa

46、y an average of $77. （令人吃惊的是，调查显示， 5564 岁之间男性的平均医疗支出是 88美元，超过 65 岁男性的该项支出是 77 美元） While noninsured expenditure rises from $66 to $80 in these age groups, insured expenditure declines from $105 to $70. （然而，在该年龄段的人的 非保险支出从 66 美元上升到 80 美元，保险支出则从 105美元下降到 70 美元） The conclusion is tempting that insurance

47、 companies are particularly wary of giving medical insurance to older people.（结论是饶有趣味的：保险公司对向老人提供保险尤为谨慎）The prin ciple of adverse select ion is pote ntially prese nt in alllines of in sura nee.The followi ng stateme nt appears in an in sura neetextbook written at the Wharton School:（逆向选择问题是所有类型的保险都存

48、在的普遍问题。下面的观点摘自沃顿商学院的一本教科书：）There is pote ntial adverse selectio n in the fact that healthy term in sura nee policy holders may decide to term in ate their coverage whe n they become older and premiums mou nt.（当逐渐变老的那些健康保险的投保人面对保费上升时，他们就有可能决 定退出保险，这种情况中便存在潜在的逆向选择）This actio n could leave an in surer

49、withan un due proporti on of below average risks and claims might be higher tha n an ticipated.（这一行动能使保险公司的客户中高风险客户所占比例过高，而使保险公司可能支付比预期要高的赔偿金额）Adverse selectio n appears （or at least is possible）whe never the in dividual or group in sured has freedom to buy or not to buy, to choose the amount or pla

50、 n of in sura nee, and to persist or to disc ontinue as a policy holder.（当保险个人或团体能自由选择购买或不买保险，能都自由选择保险金和保险计划，自由选择维持保险或中断保险时，逆向选择便会出现，或至少可能存在逆向选择）Group in sura nee, which is the most com mon form of medical in sura nee in the Un ited States, picks out the healthy, for gen erally adequate health is a

51、preco ndition for employme nt.（团体保险是美国医疗保险中最普遍的保险形式，它用于鉴别人们的健康状况，因为通常情况下，健康是被雇佣的一个先 决条件）At the same time this means that medical in sura nee is least available to those who n eed it most, for the in sura nee compa nies do their own adverse selection.（同时，这意味着保险公司自身的逆向选择行动可能使那些最需要保险的人轻易得到保险）This adds

52、one major argume nt in favor of medicare.（医疗保险市场上On a cost的逆向选择现象为支持政府医疗保障方案提供了一个主要论据）ben efit basis medicare may pay off: for it is quite possible that everyin dividualin the market would be willi ng to pay the expected cost of hismedicare and buy in sura nee, yet no in sura nee compa ny can afford

53、 to sell him a policy - for at any price it will attract too many lem on s.（通过成本收益分析，我们可以看到医疗保障是值得的，因为很有可能市场上所有人 都希望付出一定的医疗保险的期望成本来购买保险，但却没有保险公司愿意出售 保单，因为无论在那个价格水平下，保险合同的购买中都会有太多“柠檬”）Thewelfare economics of medicare, in this view, is exactly analogous to the usual classroom argume nt for public expe

54、 nditure on roads.（从这一方面来看，对医疗保障方案的经济分析与平常我们关于政府利用公共支出建设马路的 讨论十分相似）（二）少数名族就业The Lem on sPri nciplealso casts light on the employme nt of min orities.Employers mayrefuse to hire membersof minority groups for certain types of jobs.（柠檬原则还能用来解释少数民族就业问题。雇主可能在某些职位上不 愿意雇佣少数名族的职员）This decisi on may not refl

55、ect irrati on ality orprejudice -but profitmaximization.For race mayserve as a good statisticfor the applicantssocial background, quality of schooling,and generaljob capabilities.（这种决策并非是带有成见或者歧视，而是追求利润最大化的结果。对于求职者的社会背景和工作能力，种族可以是一个很好的统计指标）Goodquality schooling could serve as a substitute for this s

56、tatistic;by grad ing stude nts the schooli ng system can give a better in dicator of quality tha n other more superficial characteristics.（高质量的教育水平可以用来作为该统计指标的一个补充，教育系统通过区分学生来作为一个比其他 表面特征更优越的显示信号）As T. W. Schultz writes, The educatio nal establishme nt discovers and cultivates pote ntial tale nt. Th

57、e capabilities of children and mature students can never be known until found and cultivated. （Italics added.） （正如舒尔茨所写的那样“兴办教 育能够发觉和培养潜在的天才， 儿童和成人的才能如果不经发掘和培养将不会被 知晓”）An untrained worker may have valuable natural talents, but these talents must be certified by the educational establishment before a company can afford to use them. The certifying establishment, however, must be credible; the unreliability of slum schools decreases the economic possibilities of their students. （虽然一名未受过教育的工人可能具有十分 有价值的才能，但是