The Market for Lemons: Peaches, Phools and Phishers

The other day, a non-economist friend asked me about Janet Yellen’s husband [George Akerlof] and I was dragooned into explaining what his famous “Lemons” paper was all about.1 Then Jason Zweig of the Wall Street Journal published his interview with Ackerlof and Shiller again raising the issue of asymmetric information.2 Here is a re-think of some of the fundamentals raised by the notion of poorly informed market participants: the Phools who are Phished by competitive markets.

I missed the first edition (1970) of the Lemons paper as I had already left organized economics for the business world. In thinking about Macro most of this summer at my fire-surrounded California home, I have been focusing on what is “left out” of conventional macroeconomics. I chanced this morning to see Zweig’s note in the WSJ about Akerlof and Shiller (my former student) telling us that competition produces thievery and questionable products offered by “tricksters.” The timing of the interview, coming after the incredible disclosure by Volkswagen that it had tampered with the results of EPA pollution tests on its diesel cars underscored the “trickster” nomenclature. Score one for Zweig’s timing. Still, the underlying critique of competitive markets is a long-standing thread among social reformers and also has its place within the economics profession. The Lemons paper is part of that thread. The issue is the relevance of that model for consumer well being.

I don’t remember Adam Smith or Milton Friedman or George Stigler telling us that competition abolished all the less desirable traits of humankind. I thought their message was that the social good was promoted in spite of the knavery of the butcher, the baker and the candlestick maker. I don’t believe that the Wealth of Nations pretends to be the description of the perfect society—perfect in the sense of the much-desired Kingdom of God. Nor has the constant rant against competitive capitalism produced a demonstrably clear alternative that has stood the test of time. The absence of a demonstrated alternative must then mean that we must evaluate the success of competitive markets not as an abstraction but as a comparison to other likely systems of organizing economic activity.

Moreover, I thought the central theme of many writers since Smith was that the motives of individual market participants are to a large extent not relevant in creating a plentiful supply of goods and services that are efficiently priced. Of course, it is the word “efficiency” in the last sentence that sets the roosters to crowing that something new is feeding in the barnyard. Did Smith leave out Phools and Phishers now current in behavioral models of human action? Must we now pay close attention to such individual behavior in order to evaluate competitive market outcomes? Let me back up a bit.

Having been asked to explain the Lemons concept, I did so, but I wanted to be sure I got the essence correct. This morning, I decided to check if I had rendered the Lemons story properly in lay terms, and went back to look up the famed Ackerlof paper. I didn’t have the time to read through the original piece. Instead I skimmed the Wikipedia summary for its central finding: using the Used Car Market as the example, the contention was that the “peaches” are driven from the market, meaning that the market only consists of lemons!

Admittedly, the asymmetric information model has always bothered me—e.g. the small businessman gets an “inferior” loan deal because the banker knows less about the potential borrowers’ true state. Why? I was a small businessman and I borrowed. Was I then a Phool? Did I get taken-in by my Phisher banker, or did I search for the best deal I could find considering my own idiosyncratic circumstances? Did my banker trick me into accepting an inferior loan? Back up to the Lemon’s truck. Presumably a rational businessman checks the credit opportunities open to him. He chooses from the alternatives open at the time. It could be a Lemon, a Peach or a Prune.

What happens to the peaches in the market for lemons? They must be piling up somewhere in the universe—like the third order of smalls we keep throwing away in most empirical studies. (I always feared that the Kingdom of the delta squares and the delta cubes would somehow bring the entire universe down on our heads. Conservation of matter must have mean something. Maybe it will be a squishy with peaches instead of raindrops falling on our heads?

What struck me this morning as I feared to look at the market smashing my paper wealth to smithereens was that Wikipedia never said anything about what happened to the peaches except they were driven from the market.

Where did they go? I mooted the possibility that there were so few Peaches on the used car lots of America that the intrepid used car salesmen themselves drove them home instead of selling them off to customers. But clearly, lemons to peaches is some sort of a quality continuum. However many cars are sold annually on the used car market, I have never seen a special lot where only Peaches are sold, nor do I assume that every used car lot consists only of Lemons waiting for a Phool. Am I wrong? Maybe all of the last two decades of specialized car lots that warrant everything but the VW diesel’s actual pollution emissions, are filled only with Lemons (that of course may now change)? How can that be? One answer is an undersupply of smart used car salesmen. Somehow, that is not a satisfying answer. Something is left out of the original characterization of how the used car market works.

Moreover, as I thought about the Lemons problem, I was reminded of the similarity of standard Chicago stock-flow problem that Milton Friedman used to ask us to solve. Think about it. The market in question is for used cars—and let us ignore the fact that a used car, even a Lemon, represents a household capital item. For now, just focus on the flow market for used cars.

The cars start off largely in an auction market where new car dealers are shedding the trade-ins that they deem unlikely to sell themselves. The Used Car Mavens at the auctions rip that market apart, pricing Peaches and Lemons appropriately. The New Car Sellers have some knowledge of the value of a vehicle (whether it is Peach, a Lemon or some other kind of fruit) and their objective is to clear their own lots. (I guess they hold back some of the Peaches themselves—but if that is true the entire structure of the Lemons theory kind of falls apart). So let’s just say, be it a Lemon, a dried up Prune or a fresh, firm juicy Peach, they all show up at the auction and the omnivorous Used Car tribe devours each of them at whatever price they can get away with.

For a trade to occur, there must be a willing seller and a willing buyer—both somewhat knowledgeable—although in this case the Used Car tribe must be at somewhat of a disadvantage themselves, since they have but a short time to inspect the available vehicles which represent only the ones the new car dealers don’t wish to carry on their own lots. Still, the market clears.

Now, the collage of fruit shows up on the Used Car Lots of America. Fearful buyers (the Phools) begin to shop, in terror that they will be stuck with a Lemon. According to the original version of the theory, the Peaches have disappeared. Competitive equilibrium thus leaves us with a highly inferior outcome. Only the Lemons survive!

That’s what still bothers me. We are purportedly measuring a flow demand curve and supply curve, but the observed supply curve is missing all the Peaches. How can that be a sustainable equilibrium? The theory says that the actual price paid is some sort of a combination of Peach and Lemon prices—but then the Peaches go somewhere—and all that is left are Lemons. This is clearly an incomplete characterization of how the used car market works today. I doubt it worked that way in 1970.

The Lemons theory without an adequate specification of true stock-flow equilibrium, throws a spanner into the works. The model is incomplete in a crucial way and welfare implications drawn from such a model are likely to be highly suspect in the best case, totally erroneous in the worst case.

The Lemons problem, structured as it was in those days, was a kind of incomplete game. Yet, a flow market is clearly a repetitive game in which learning must be taking place on both sides of the market. Suppliers gather information from yesterday’s outcome and buyers learn about faulty transmissions dropping out of their neighbor’s used cars. Thus, any inference drawn from the actual observed prices of used cars must somehow incorporate that prior learning which goes on day by day. Indeed, that explains the evolution of the Used Car Market today—a huge change from the mid 1960’s. Now reputational capital supplied by national used car dealers—public companies now as well—is involved in an efficient pricing game, with warranties and ‘take back’ policies available to a complaining Lemon buyer. And, as far as I can tell, the stock of Used Cars doesn’t have a special lot only for Peaches either.

Dynamic Stock Flow equilibrium seems to sharply weaken the asymmetry case. Learning behavior is absolutely critical to understanding the behavior of market participants. Yesterday’s Lemons are not just thrown away. Somehow, they get incorporated into the pricing of currently available used cars. Their values rise and fall upon much more comprehensive knowledge about the performance of various models and makes and vintages. Dealers who specialize in selling Lemons to Unwitting Buyers must achieve something of a reputational equilibrium as well. Why would we think the Buyers who talk among themselves over time, so to speak, don’t know which dealers have better fruit and what the price of damaged fruit ought to be?

Still, however, we get the assertions from very learned and honored economists who claim the competitive model leads to flourishing thieves who prosper while Phools suffer. No one seems to learn. Common sense tells us that the market is made up of both—and all the rest of us in between.

Yet, pricing for quality works itself out. Better quality goods get better prices over time than poorer quality goods, while one can always find both kinds of fruit in any market. Cheap imitations of first quality goods sell for significantly lower prices. How is that possible? And how is it possible that new suppliers of varied quality enter markets with questionable quality goods—only to be sorted out by repetitive buying and selling? Apparently, even Phoolish buyers get to taste the fruit from time to time. Learning is essential in all markets over time. Nothing surprising here.

Under most conditions, the second golfer on the green, who puts last, has more information than the first to put. Learning by doing and reputational capital are permanent fixtures of all efficient markets. Otherwise, we would never get anything like an approximation to stock-flow equilibrium. Confusing an individual action and long-term market equilibrium is excusable for non-economists. Should it be for Nobel Prize winners?

PT Barnum was right, partially. A sucker is born every minute and by definition Suckers lose money. Others learn, however from those unfortunate outcomes. Over time, some of those suckers learn from their earlier mistakes and other market participants learn as well. The story doesn’t end here. The issue of who’s the fish, the sucker in the game is relevant. Are there other ways to organize trade and production that would alleviate this problem?

We need to ask, what alternatives are there to providing adequate information to markets filled with Phools and Phishers? The Government? Regulations? Here comes the Saint of non-market solutions. As far as I can tell from many years dealing with many governments in many countries, government gets its share of Phools as well. Worse, the compelling power of Law and Regulation often enshrines faulty views that are hard to extract from the regulatory complex. Furthermore, any given set of new regulations will send signals to market participants to find a work-around. The “regs” themselves set up a system of rewards and punishments that create incentives for “bad” and “good” behavior. In a public market context, this means that complex public companies have a manifold set of departments whose employees will be given inherently conflicting goals: “keep the product priced so we can sell it,” “expand output so that we can maximize profits,” and “stay within the regulations.” These are very complex instructions involving lots of tradeoffs between the letter of the law and the compensation available for those who don’t get caught.

Since we now tend to pay employees who produce the best results, we create incentives to find a “work-around ” within any given regulatory structure. Markets also tend to create larger and larger enterprises because of supposed economies of scale and scope. This leads to a complex set of organizations whose activities fall into “silos.” We then expect our senior executives and the Boards that supposedly supervise those managers to maximize shareholder value and stay out of trouble, presumably by looking down each silo to see what is going on all the way down. That creates a never-ending scenario of risk and return within which the various activities of a corporation get organized under a common capital base. Unavoidably, “solutions” occur that appear to solve the problem yet bury the inherent risk. A shock occurs and the risk certainly appears and we ask, “How did that happen?” Even a scrupulous honest senior manager cannot just look down into the silos and spot all deviations from “best practices.” We then get a “London Whale” or the “VW Fraud”—not for every enterprise, but for enough to keep journalists well paid to investigate possible If the final outcome were only entertaining investigative journalism that discloses what is not clearly obvious to market participants, the outcome would be tolerable. Sadly, we get more. Some are so offended that they go to the Government to regulate and get rid of the apparent problem recently disclosed. Some regulations do improve market performance. But, if we expect the Government regulators and legislators to be totally observant of the advice given by Caesar when he divorced Pompeia, who are the Phools?

PT Barnum was right…but the real suckers are those that believe that by saying Government or more regulations will solve such problems efficiently or in a timely manner. The evidence is in. Even regulation doesn’t avoid Lemons, even if changes where they appear.



  1. Akerlof, George A. (1970). “The Market for ‘Lemons’: Quality Uncertainty and the Market Mechanism”. Quarterly Journal of Economics 84 (3): 488–500 []
  2. Zweig, Jason “Phishing for Phools’: A Q&A With George Akerlof and Robert Shiller, WSJ Jason Zweig 9/215/2015 []