# What are some results in Economics that are both a consensus and far from common sense?

What are some results in Economics that are both a consensus between most economists and far from common sense?

I would also welcome suggestions of clear definitions for what we should mean as consensus , specially considering that economics is an area with a lot of methodological divergence. Let me try first, a suggested definition for consensus in this setting would be:

the existence of a group of experts that would claim that the result is certainly true.

• I was expecting to see an answer with "markets are efficient". I'm honestly unsure if it's not there because of a lack of consensus or not there because economists have unusual notions of common sense. – psr Mar 13 '18 at 18:59
• @psr Because "markets are efficient" only if all the requirements for the welfare theorems to work are met. Introduce externalities, private information, transaction costs, fix costs, etc. and suddenly you end up with something not efficient at all. In most situations not all of those requirements are met. But in a lot of situations it is also hard to do better than the market. Then there a bunch of situations where you can internalize externalities with some intervention, etc. So it really depends on the situation – Felix B. Mar 14 '18 at 8:14
• The law of unintended consequences is common knowledge in economics. It is not exactly far from common sense, but it is very often (incl. willfully) forgotten or ignored in economic (or pseudo-economic) analysis. I understand this law to mean that any coercive or involuntary economic action is incrementally destructive; if the action was taken under the pretext of a common good, it will have unintended consequences. There are other, similar formulations of this law. – Jake Mar 14 '18 at 18:45
• @Jake -- It's a great question. The dissonance surrounding the Monty Hall paradox is the distinction between Monty picking a door at random, or doing so with knowledge. Monty knows which door has the cash, and deliberately chooses the door that does NOT. You choosing one of the other doors forces Monty's hand, and thus changes the equation. It becomes easier to grasp if you change it to 100 doors, you pick one, and Monty opens 98 of them with worthless prizes.... – Stephen R Mar 14 '18 at 20:37
• @FelixB.The less complete information is, the more attempting to improve efficiency with central planning fails, because the planners themselves have less information than the market in aggregate has. Prediction markets have repeatedly shown themselves more accurate than the best experts. – Monty Harder Mar 14 '18 at 21:22

As Paul Samuelson (1969) put it:

thousands of important and intelligent men ... have never been able to grasp the doctrine [of comparative advantage] for themselves or to believe it after it was explained to them.

Example

Imagine that an American worker who devotes all his time to soybean production can produce up to 100 tons of soybeans per year. And if he devotes all his time to steel production, he can produce up to 4 tons of steel per year.

In contrast, the corresponding figures for a Chinese worker are 30 tons of soybeans or 3 ton of steel.

Maximum possible production

American  Chinese
Soybeans     100      30
Steel         4        3


A layperson could reason:

An American worker is literally more productive than a Chinese worker at everything. So why aren't we simply producing all of our own soybeans and steel?

Instead, we're doing the foolish thing of importing steel from China!

This reasoning is "common sense". It is also wrong.

Although the American worker is "better at everything" (we say he has the absolute advantage in producing both soybeans and steel), the Chinese worker has the comparative advantage (CA) in producing steel. This is because by producing 1 ton of steel, the American forgoes 25 tons of soybeans, while the Chinese forgoes only 10 tons.

And so, by the principle of CA, the American should focus on producing soybeans and the Chinese on producing steel. The two can then trade to mutual benefit.

Numerical example:

Say that without trade, the American spends a quarter of his time producing steel and the rest producing soybeans. The Chinese spends half his time on each. Hence:

1. Consumption without trade

American  Chinese
Soybeans     75       15
Steel         1       1.5


But they can do better by specializing and trading. The American, whose CA is in soybean production, should specialize in soybeans. And the Chinese, whose CA is in steel production, should specialize in steel.

2. Production after specialization but before trade

American  Chinese
Soybeans     100       0
Steel         0        3


The American can then trade, say, 20 tons of soybeans for 1.2 tons of steel. End result:

3. Consumption after specialization and trade

American  Chinese
Soybeans     80       20
Steel        1.2      1.8


Comparing Scenarios #1 and #3, we see that with specialization and trade, both the American and Chinese workers are strictly better off. Remarkably, each gets to consume more of both soybeans and steel than they did without trade.

Thus, even though the American is "better at everything", the principle of CA offers a powerful rationale for why he should still import steel from China and be "dependent" on the Chinese worker.

Most theorems in economics would satisfy the consensus requirement. However, depending on what you consider to be common sense, different results will qualify. The following are two results that I found sufficiently hard to believe when I first encountered them.

The revenue equivalence theorem, which, according to Wikipedia, implies that

any single-item auction which unconditionally gives the item to the highest bidder is going to have the same expected revenue.

Arrow's impossibility theorem, which, according to Wikipedia, suggests that

no rank-order electoral system can be designed that always satisfies these three "fairness" criteria:

• If every voter prefers alternative X over alternative Y, then the group prefers X over Y.
• If every voter's preference between X and Y remains unchanged, then the group's preference between X and Y will also remain unchanged (even if voters' preferences between other pairs like X and Z, Y and Z, or Z and W change).
• There is no "dictator": no single voter possesses the power to always determine the group's preference.
• It is a bit of a stretch to call Arrow's impossibility theorem a result in economics... – user16614 Mar 14 '18 at 23:03
• @Servaes: Why not? The result was first published in the Journal of Political Economy, an economics journal, and Arrow worked on the problem while completing his PhD in --- guess what --- economics. – Herr K. Mar 14 '18 at 23:37
• Without additional context, the description of the "revenue equivalence theorem" seems to imply that auction prices are independent of the item being auctioned, e.g. that a single-item auction for a toothpick would have the same expected revenue as an auction for a yacht, along with other absurdities. Presumably these absurdities are prohibited, but if so, what seemingly absurd conditions are allowed such that the theorem's counter-intuitive? – Nat Mar 15 '18 at 3:45
• @Servaes: Moreover, Arrow's theorem concerns the problem of preference aggregation, which is well within the scope of economics. – Ben - Reinstate Monica Mar 15 '18 at 4:05
• @Acccumulation Just so I don't come off the wrong way, not trying to be excessively pedantic here or anything. My concern was just that, by itself, the quote's very misleading. That theorem's true only in very idealistic models. But since this is a question about theorems that seem counter-intuitive, it'd be nice to have a statement of when it does apply, to highlight how it might be counter-intuitive in those cases. I mean, it seems highly counter-intuitive in the general case, but that's actually good because it's not true in the general case. – Nat Mar 16 '18 at 20:22

In an open economy, the balance of payments current account equals net saving. This is often represented as:

$$S - I = X - M$$

where $S$ is saving, $I$ is investment, $X$ is exports and $M$ is imports. That is admittedly a slight over-simplification as current account includes not only exports and imports of goods and services but also other items such as income from foreign investments or employment abroad, and foreign aid. For many countries, however, the net amount of these other items is relatively small so that the balance on trade in goods and services approximates fairly closely to net saving.

This appears to diverge from common sense since, if a country has a trade deficit, most non-economists looking for explanations will consider possibilities such as:

• lack of competitiveness of domestic firms;
• 'dumping' by foreign producers;
• an overvalued exchange rate.

Very rarely will a non-economist suggest that a trade deficit has anything to do with levels of saving and investment.

Note that saving and investment here include those of both the private and the government sectors. So one implication of the above is that a government deficit, unless offset by private sector net saving, will be associated with a trade deficit (just 'associated' because the direction of causation is a further question).

• It seems common-sensical enough that the trade surplus - being the amount of money your country has made from selling stuff that it hasn't yet used to buy stuff - would be equal to the total savings in the country since on an individual level that is the definition of savings... – user253751 Mar 14 '18 at 4:40
• Also all of the above factors would seem to ultimately relate to income and expenditure, which relate to savings. – user253751 Mar 14 '18 at 5:21
• This only seems to mean that countries with a trade surplus will have to put this surplus somewhere ... Moreover, this balance theorem doesn't answer the cause of trade deficit, while the 4 allegedly stupid explanations do. – agemO Mar 16 '18 at 10:12
• @curiousdannii Done! – Adam Bailey Mar 16 '18 at 11:08
• @agemO I'm not suggesting that the 4 explanations are stupid, only that they are not the whole picture. – Adam Bailey Mar 16 '18 at 11:10

The Giffen Paradox - increasing prices can lead to higher demand even if the goods are considered inferior.

General consensus is increasing prices lead to less demand - if it's more expensive, people will buy less.

In some cases, increasing the price will make consumers perceive a good to be of higher quality, or more desirable, thereby increasing demand. (Example - if iPhones cost only one third of what they do, nobody would spend their money on a phone that doesn't even run Android).

However, even with inferior goods, rising prices can lead to higher demand. This paradox was first observed by Giffen in the 19th century, when rising potato prices meant poorer people weren't able to afford the occasional egg or piece of meat anymore, buying more potatoes instead.

• This could possibly be called an "observation" more than a "result", but it's something that's easy to comprehend once you read the explanation, but hard to wrap your head around without an example. – Guntram Blohm supports Monica Mar 13 '18 at 18:02
• Depending on who you talk to, an iPhone might be a perfect example of an inferior good ;) – curiousdannii Mar 14 '18 at 13:45
• This is similar to the response of thrifty savers to falling interest rates: "We better save more to compensate!" Which thwarts political efforts to stimulate spending. – MarkHu Mar 14 '18 at 18:08
• For many years the real world evidence of Giffen goods was very poor. The evidence from potatoes in the 19th century is discredited. I believe Battalio, Kagel , Kogut (1991) are the first to provide hard evidence, from experiments. We didn't get real world evidence until Jensen and Miller (2008). Most people who believe in Giffen goods probably do so based on low quality or falsified "evidence". – BKay Jan 25 at 2:57
1. The fact that the burden of a tax on sellers can by borne by buyers, and vice versa. More generally, the fact that true tax incidence is largely or completely unrelated to who is nominally being taxed (e.g. taxes on yacht purchases can in principle hurt the poor more than the rich, etc.).

2. The fact that in a perfectly competitive market with free entry and exit, all firms make zero profits in the long run (if opportunity costs are taken into account).

3. Coase's theorem: "if trade in an externality is possible and there are sufficiently low transaction costs, bargaining will lead to a Pareto efficient outcome regardless of the initial allocation of property". (E.g. under a cap-and-trade system for pollution permits with sufficiently low transaction costs, the final allocation of permits is independent of the initial allocation, even if some permits are sold and others are arbitrarily given away for free.)

4. This one's a bit more "in the weeds", but the difficulty of eliminating the marriage penalty: "it is mathematically impossible for a tax system to have all of (a) marginal tax rates that increase with income, (b) joint filing with income splitting for married couples, and (c) combined tax bills that are unaffected by two people's marital status."

• I am guessing because sellers can 'pass on' the cost of taxes to their buyers via increased price, with equivalent processes working the other way, and this passing on of cost propagating through an economy is how yacht purchase tax could theoretically land on people far removed from yacht buyers? – benxyzzy Mar 14 '18 at 12:03
• @benxyzzy I wasn't thinking it would land on people "far removed" from yacht buyers, but rather on yacht manufacturers. If the demand for yachts is much more elastic than the supply, then a large tax on yacht purchases would cause the before-tax price to drop much further than the after-tax price rises, so that the wealthy purchasers are only affected modestly, but the (probably poorer) manufacturers get hit really hard. – tparker Mar 14 '18 at 18:03
• I'm calling foul since this proposes 4 different ideas, with different levels of detail. Also, #2 is especially arguable since it is about an impossibly idealized theoretical situation populated by mindless actors (i.e. in the real world, not all actors choose to enter all profitable markets, and the smart early-entry firms should not stick it out in the face of decreasing-profits.) – MarkHu Mar 14 '18 at 18:03
• @JollyJoker Sales taxes hurt yacht manufacturers whether or not they're locked into the industry. If they choose to keep selling yachts, they're hurt by the forgone sales and lower after-tax sale price. If they choose to go into other industries, they increase the supply of labor in those other industries and therefore push down the wage. The same argument holds for buyers, with "supply", "down", and "wage" replaced by "demand", "up", and "price" respectively. If this weren't the case, then the government could use targeted sales taxes as a free lunch for raising revenue without harming anyone – tparker Mar 15 '18 at 17:05
• #4 isn't in the "counterintuitive but true" category; rather, it's in the "summary is false, full explanation is obvious" category. It's easy for a tax system to eliminate the marriage penalty, either by allowing married partners to file as if they were both single, or by setting the tax bracket thresholds for married couples at double the corresponding thresholds for single people. – ruakh Mar 17 '18 at 6:26

Best price clauses and, to a lesser extent, price-matching guarantees have been the subject of intense regulatory activity in recent years. Here's a fact that is surprising to many, despite there being a significant consensus in the economics profession:

Best price clauses and price-matching guarantees can harm competition and consumers

A best price clause/most favored nation clause/price-parity clause requires a seller listing a price via one venue (e.g. a price comparison website) to ensure that the price listed there is no higher than that available through other similar venues. It is often imposed by venue operators to ensure their venue will attract customers. Common sense suggests that a clause requiring a vendor to be offered the lowest available price should be, at worst, neutral for consumers.

Suppose there are two venues, $A$ and $B$, that a seller can sell through. Suppose both are large suppliers of business to sellers, so simply quitting one of the venues is not a viable option.

Here's the problem: if venue $A$ charges a commission of $c_A$ to sellers who sell through its platform and venue $B$ charges commission $c_B>c_A$ then sellers will set a lower price on $A$ than on $B$ to try to steer consumers to buy through venue $A$ (where it pays lower commission). Thus, a venue can have consumers steered towards it by cutting its commission--venues compete in commission. Moreover, lower commissions (which are essentially a variable cost for the seller) are passed-through to consumers in the form of lower prices.

Now suppose there are best price clauses in effect. If $c_A<c_B$ then sellers can't steer consumers towards $A$ by setting a lower price on $A$ because $B$'s best-price clause requires the price on $B$ to be no higher than that on $A$. Thus, $A$ can no longer attract customers by cutting its commission, and so has no incentive to compete in its choice of $c$. This results in higher $c$s and higher consumer prices. This effect is theoretically robust and empirically well-validated.

A price matching guarantee is a promise from a seller to a consumer of the form "If you find the same product at a lower price elsewhere, I'll beat that better price". Common sense suggests a guarantee to beat the lowest price in the market should be good for consumers. Not necessarily so.

Here's a rough illustration of why: Suppose sellers A and B both have price-matching guarantees and consumers have a preferred seller from whom they buy by default unless the other seller offers a better deal. Normally, sellers would cut their prices to try to attract consumers from their rival. But here that doesn't work! If A cuts its price then the consumers whose default is B can just go to B and get it to match A's reduced price. But this means A has no benefit from lowering its price and will just stick with the same high price that it had all along. The price-matching guarantee has completely killed price competition!

• Isn't this exactly common sense? Why else would sellers use price matching guarantees rather than actually lowering prices? – mattdm Mar 17 '18 at 14:45
• @mattdm Well, one could easily tell a story about consumers who face search or shopping frictions and therefore can't be sure where the lowest market price is (or firms that can't perfectly monitor rivals' prices and sometimes end up charging more than rivals by mistake). Such a guarantee would then plausibly be beneficial for consumers. In any case, it was a big battle for economists to get competition policy makers (who are more informed than the general public) to see the potential equilibrium harm of these things, suggesting that it's not that common sense. – Ubiquitous Mar 17 '18 at 18:51
• If the price is the same at A and B then, unless there's a pretty big difference in customer service etc, I'll just buy the product from whichever one I happen to visit second. Which, actually, might be the worse one, if I'm in the habit of checking my preferred store first. Otherwise I have to go back to the first store. – user253751 Feb 11 at 15:43

Lowering income tax rates can in some circumstances increase revenue. A simplistic view would assume higher taxes = higher revenue, but it doesn’t account for the fact that different tax rates alter behavior. It becomes obvious when you look at the extreme end — with a 100% income tax rate, nobody would bother having a job, because they don’t get to keep any of their wages. Government income from the tax would plummet.

The overall concept is described in a theory called The Laffer Curve

• I don't think this is far from common sense. – user253751 Mar 14 '18 at 0:48
• I’ve often encountered people who believe the concept of the Laffer Curve to be false; that obviously higher tax rates increase government revenues. Pay a bit of attention to politics and you’ll see it quite often – Stephen R Mar 14 '18 at 0:51
• The interesting question here is whether the "some circumstances" actually exist anywhere at present. If the circumstances don't exist, i.e. no actual current taxes can be reduced to increase revenue, then the theory is simply an abstract curiosity and common sense doesn't come into it. – Mike Scott Mar 14 '18 at 10:54
• The basic principle of the Laffer curve is irrefutable. However, I'm skeptical that there is a society anywhere that is actually on the high end of the curve (i.e. that would increase revenue by lowering taxes). Even Ronald Reagan's fans don't seem to talk about the Laffer curve much anymore. – Kef Schecter Mar 17 '18 at 5:25
• It worked for Reagan just fine. We’ll see if it works for Trump; but the fact that companies are repatriating billions of overseas dollars suggests it might. (The difficulty administrations seem to have is to not increase spending at the same time, but government revenues definitely went up under Reagan.) – Stephen R Mar 17 '18 at 12:45

I’m going to toss my hat into the ring with the very notion of

# Opportunity Cost

Who hasn’t argued with someone who “likes” A, which has an opportunity cost B, and have that person adamantly refuse to consider the loss of B as a cost for A or in any way relevant to the choice to get A? If you’ve ever tried to persist in such a situation, I’m sure you also quickly found the other person getting offended, as if you were insulting A and/or them for liking A.

A large part of the problem is, of course, the not-so-subtle verb substitution I used: the person in question here is talking about, and thinking about, “liking” A. The speaker, you or I in the hypothetical, is instead talking about the decision of whether or not to procure A. The argument is, fundamentally, that A can be good, worth “liking,” but not worth it—because of the opportunity cost B. And if anyone knows of a good way to explain that and assuage hurt emotions over feeling attacked for liking A, I’m all ears! Recognizing the problem does not itself produce a solution to it.

Anyway, unlike a lot of other issues on this page, which are important but not a daily concern, opportunity cost is, or at least could be, at the heart of just about every decision every person makes. It is relevant to all people, but a lot of people seem to not just ignore it, but to actively disdain the very concept. Therefore it seems, to me, like a strong contender here.

• See: Ferraro Paul J & Taylor Laura O, 2005 "Do Economists Recognize an Opportunity Cost When They See One? A Dismal Performance from the Dismal Science". The B.E. Journal of Economic Analysis & Policy, De Gruyter, vol. 4(1), pages 1-14, September. (The paper is widely available and discussed elsewhere on the internet. 78% of economists gave the wrong answer to a basic multiple choice question on opportunity costs, although the wording of the question has been criticised by others.) – Silverfish Mar 18 '18 at 1:43
• The question they posed was: "You won a free ticket to see an Eric Clapton concert (which has no resale value). Bob Dylan is performing on the same night and is your next-best alternative activity. Tickets to see Dylan cost 40 dollars. On any given day, you would be willing to pay up to 50 dollars to see Dylan. Assume there are no other costs of seeing either performer. Based on this information, what is the opportunity cost of seeing Eric Clapton? (a) 0 dollars, (b) 10 dollars, (c) 40 dollars, or (d) 50 dollars." – Silverfish Mar 18 '18 at 1:44
• @KRyan: Can I suggest you delete your comment so that others who wish to contemplate the problem on their own before looking up the answer are not influenced by your answer? There's no need to ask here if you're correct; you can find the correct answer yourself, plus the reasoning behind it, in the paper. – cjs Mar 18 '18 at 10:43
• @Curt Sure, but no I could not have done that, because Silverfish’s link wanted to charge money to access the paper. – KRyan Mar 18 '18 at 13:52
• It's always worth trying a quick web search. That's how I got the link to a free copy in my comment above. (I don't know how long it will last, which why I'm mentioning here how I found that link.) – cjs Mar 18 '18 at 17:51

Price discrimination can make consumers better-off

This can happen in a variety of ways. For example, suppose that 50% of consumers are 'loyal' to firm A and 50% loyal to firm B. A consumer is willing to pay $v$ for one unit and incurs a switching cost $s\in(v/2,v)$ if they buy from a firm other than the one to which they are loyal. Consumers buy from the firm offering the highest utility, breaking indifference in favour of the firm to which they are loyal.

If firms must offer the same $p$ to all consumers then, the only pure strategy equilibrium has $p_A=p_B=v$.

If we allow firms to price discriminate by offering different prices to loyal and non-loyal consumers then $p_i=v$ is never an equilibrium and there is an equilibrium in which $p_i^{\text{loyal}}=s$, $p_i^{\text{non-loyal}}=0$.

Without price discrimination firms don't compete to attract the rival's consumers. Intuitively, it takes a price cut of at least $s$ to do so, but that means also losing $s$ units of profit from loyal consumers who also benefit from the price cut. With price discrimination, on the other hand, a firm can fight for its rival's loyal without sacrificing any profit on its own. But this goes in both firections so both firms end up having to fight for their loyals against a rival that is trying to poach them.

Another ways that price discrimination can benefit consumers is by giving a firm enough profit to make a product viable (otherwise the product wouldn't get produced at all, leaving consumers with zero surplus).

There is the monetary Impossible trinity concept in international economics:

The impossible trinity (also known as the "trilemma" ...) is which states that it is impossible to have all three of the following at the same time:

• a fixed foreign exchange rate
• free capital movement (absence of capital controls)
• an independent monetary policy

Granted there may not be much "common sense" thinking on such an esoteric concept. But the Economist magazine thinks it is pretty important according to https://www.economist.com/news/economics-brief/21705672-fixed-exchange-rate-monetary-autonomy-and-free-flow-capital-are-incompatible . Also, more people are becoming aware of such things since the blockchain and cryptocurrency popularity surge in late 2017. Also,

• Common sense tells me you can't have #1 and #3. Then you might tell me about capital controls and I might go "oh yeah, I guess that would kinda work, though it causes its own problems." – user253751 Feb 11 at 15:47

Let's say current engine technology allows vehicles to go one mile on a gallon of gas. Improving engine technology to achieve 10 MPG must decrease the amount of gas used, right?

Actually it depends. Increased efficiency has two effects. One effect is that per unit of distance, less gas is consumed. Another effect is that driving one unit of distance is now cheaper, increasing travel. How much travel increases depends on how much the demand curve rises moving from the old unit price to the new unit price. In fact travel can increase by a larger factor than efficiency increases, leading to more fuel consumption.

• It would be helpful if you could include in your answer a link to a relevant source. I was aware of this "paradox" but unaware that it had been named after Jevons. – Adam Bailey Mar 16 '18 at 19:08
• This is conceptually related to the notion of risk compensation. – Ubiquitous Mar 17 '18 at 9:08
• @Ubiquitous Yes, it is, and both profoundly affect public policy... – Reinstate Monica Mar 17 '18 at 15:35

The fact that you can correct for externalities with (tradeable/marketable) permits seems to be a consensus with economists. But considering the existing applications, the theory appears to be not common sense.

Example:

The EU/countries in the EU have implemented a Carbon Emission Trading scheme while nations within the EU continue with other regulation like promotion of solar/wind, etc. When the point of a permit trading scheme is for the market to allocate the efforts to that place where you can save CO2 most efficiently. The added regulation only distorts the permit market without actually helping to reduce the CO2 produced, as the amount of CO2 produced is set by the amount of permits issued. And by funding solar, you just make permits cheaper and increase pollution in another sector.

Another quirk is, that permits are grandfathered to companies to keep the prices down, when the gifted permits result in an opportunity cost of using them (since you could sell them at the same price), making the price rise at just the same rate as if the company had bought them. The only result of grandfathering being that it gifts money to established companies, distorting the market.

If the amount of issued permits were to high this would still be an example of policy makers not understanding the theory of permits. The thing is, either you base your CO2 reduction policy completely on permits or not at all. If you issue too many of them, then they don't have an effect and the only reduction comes from other measures so the implementation of permits is bad. But if they are actually binding (i.e. there are less permits than people want to produce) then all the other measures have no effect as they only shuffle around where the CO2 gets produced. So the implementation of other measures is bad. In either case there seems to be a lack of understanding of how permits work.

And I also want to argue that there are not too many permits anymore, the surplus was due to the crisis in 2008. And while I would still argue that the amount issued is still too high, the price is high enough above zero, that a reduction of CO2 in one place would probably lead to an increase in a different place. But again, it does not really matter. The issue is, that they did not implement a theory in a way that it actually works properly.

• This answer might be clarified by using a different term than "certificate" --that word does not appear in the article at en.wikipedia.org/wiki/Carbon_emission_trading wiki page. Maybe simply "permits," or more descriptively "regulatory permits and fees." In which case it becomes apparent that this so-called "market" for carbon is not much different than any other system of regulation via fees and/or fines. The main feature being the explicit waiver "price" (penalty.) – MarkHu Mar 16 '18 at 19:44
• Isn't the standard economists' terminology for this "marketable permits" (or "tradable permits")? And the popular name "cap and trade"? – Adam Bailey Mar 16 '18 at 22:43
• The example seems to be flawed. The amount of CO2 produced is not set by the amount of permits issued in the ETS: there's a massive surplus of permits, hence their rock-bottom price. The amount of CO2 produced in regulated sectors is significantly less than the amount of permits issued. – 410 gone Mar 18 '18 at 19:44
• Maybe the economist's "consensus" is wrong, and the commoners are right. @EnergyNumbers is correct to question how closely the artificial "market" can mirror the real world. Assuming cooperation from all actors is a stretch, depending on how antagonistic they become after the new rules are imposed on them. – MarkHu Mar 19 '18 at 21:30
• what do you mean by "cooperation from all actors"? If you don't buy permits when producing CO2 you will have to pay a penalty. So as long as the penalty is above that price it does not make sense to "not cooperate" – Felix B. Mar 20 '18 at 20:30

The "single monopoly profit theory" is often viewed as quite counter-intuitive:

Leverage of market power cannot be used to profitably foreclose a rival.

Suppose there are two products, A and B. A is monopolised and produced only by firm 1; B is supplied competitively by both firm 1 and firm 2. Common sense presents to following concern: firm 1 might try to use its market power in A to become a monopolist in B and foreclose competition from firm 2. One way to do this would be to bundle A and B1 together. Everyone who buys A would also be forced to buy B1, even if B2 were the better product. This would make it difficult or impossible for 2 to achieve any sales.

This logic is flawed, as the Chicago school pointed out. Suppose consumers will pay $v$ for A, $v$ for B1, or $v+\Delta$ for B2 (so $\Delta$ is firm 2's quality advantage). Suppose that A and B1 are bundled and that consumers buy the bundle. In a desparate attempt to win business, 2 will cut the price of B2 to zero (this is the usual Bertrand logic). Thus, consumers will buy the bundle if $$2v-p_1\geq v+\Delta\implies p_1\leq v-\Delta.$$

Thus, the best 1 can do through bundling is to earn $v-\Delta$ from each sale of A and B1. But it could do better simply by selling A at a price of $v$ and giving $B_1$ away for free. Thus, the claim that bundling/leverage of market power is a profitable way to foreclose competition turns out to be logically flawed.

Addendum: subsequent work has shown that leverage of market power is possible in various situations. But the conditions needed for it to work are more intricate than suggested by common intuition.

The assumption of the rationality of the agents is a consensus on neoclassical economics and other schools, that is often badly translated to real world examples by economists, due to the difficulty of measuring utility.

The dictator game translates this idea quite clearly: game theorists some economists claim humans often act “irrationally” when making offers larger than zero. Their mistake is to assume the utility outcome of the player is equal to the monetary outcome.

• I am a game theorist and I do not claim that. Nor would most game theorists I know. – Giskard Mar 17 '18 at 7:25
• Edited. Happy to hear that - although not the same in the 3 universities that I've studied. Even if they wouldn't write it, this would be often claimed in classes and discussions. – JoaoBotelho Mar 18 '18 at 23:44
• Thanks. I think most professors would probably make the same nuanced statement that I would make: If you accept the assumption that people want to maximize their payment then according to this model they act irrationally. – Giskard Mar 19 '18 at 8:02

This is a "less-known paradox," usually put as a counter to famous Bertrand paradox. It is a starting point in the literature on informational frictions in consumer markets, and the scientists in the field agree on its significance.

Its idea is diametrically opposite to that of Bertrand. Consider the following simple example. There are $2$ firms which produce homogeneous goods at zero marginal cost and compete in prices, $p$. This simultaneously set prices. Also there is a single consumer who supplies a demand given by $1-p$. Importantly, the consumer does not observe prices set by firms and, therefore, needs to search for them sequentially, where search is costly. Suppose that cost of visiting a firm is given by $0 < c \leq \frac{1}{2}$. Then, the unique equilibrium of the market is that both firms charge monopoly price $$p^M= \frac{1}{2}.$$

This is a diametrically opposite result to that of Bertrand.

The reasoning behind the result is as follows. Suppose both firms charge $p=0$. Then, the consumer randomly visits one of the firms, say firm $i$, and buys. However, firm $i$ could have charged $c$ and made positive profits as the consumer would have bought goods anyway because she would have suffered cost $c$ had she left firm $i$ in order to buy from the rival firm. By the same argument, one can see that $p=c$ cannot be an equilibrium as now firm $i$ can charge $c+c$ and improve its profit. Continuing this way, it is easy to arrive to an equilibrium where both firms charge $p^M$. A firm does not want to charge $p^M+c$ simply because its profit is maximized at $p^M$.

Formal Analysis of the Example

Timing: First, the firms simultaneously set prices. Second, the consumer without knowing prices engage into sequential search. The first search is free and the consumer visit each firm with equal probability. The consumer can come back to the previously searched firm for free. The consumer has to observe a price of a firm to buy goods from that firm.

Beliefs: In equilibrium, the consumer has correct belief about strategies of firms. If, upon visiting a firm, she observes a price different from an equilibrium one, the consumers assumes that the rival firm has deviated to the same price too. Thus, the consumer has symmetric (out-of-equilibrium beliefs). Note: the results of the game does not change if the consumers has passive beliefs.

Strategies: Strategies of the firms are prices. As mixing is allowed, let $F(p)$ represent the probability that a firm charges a price no greater than $p$. Strategy of the consumer is whether to search for the second price, upon observing the first one. This strategy is given by a reservation price $r$, such that upon observing a price lower than $r$ she buys outright, upon observing a price greater than $r$ she searches further, and upon observing a price equal to $r$ she is indifferent between buying immediately and searching further.

Equilibrium Notion: Concept of Perfect Bayesian Equilibrium (PBE) is employed. A PBE is characterized by price distribution $F(p)$ for each firm and the consumer's reservation price strategy given by $r$ such that $(i)$ each firms chooses $F(p)$ that maximizes its profit, given the equilibrium strategy of the other firm and the consumer's optimal search strategy, and (ii) the consumer searches according to the reservation price rule $r$, given correct beliefs concerning equilibrium strategies of firms.

Theorem: For any $c>0$, there exists a PBE characterized by triple $(p^M, p^M, r)$, where $p^M$'s are charged with probability $1$ and $$r=1.$$

Proof: First, I prove that $r=1$, or that the consumer buys outright when she observes any price lower than $1$. Clearly, if she observes a price greater than $1$ she does not buy from that firm as this yields a negative payoff to the consumer. Now, suppose she observes price $p'<r$. Then, she expects the rival firm to charge $p'$ too. Thus, if she buys outright her payoff is $\int_{p'}^{1}(1-p)dp$, and if she searches she expects a payoff equal to $\int_{p'}^{1}(1-p)dp - c$. As the former is greater than the latter, she better-off when she buys immediately. This proves that $r=1$.

Next, I prove that both firms charge $p^M$. Clearly, firms never charge above $1$ as they will never sell. Then, the expected profit of a firm is $\frac{1}{2}(1-p)p$ because the consumer visits a firm half of the time. It is easy to see that the profit is maximized at $p^M$. QED.

• @denesp I'm afraid there is no textbook that I'm aware of where the paradox is discussed. You can find short analysis of its some forms in papers on search. – Green.H Jul 19 '18 at 10:47

Nice question! I would like to add a few elusive ones that I think are important.

Natural selection does not require intelligence. It is more of a biology theory, but its economic implications are vast. Evolution of cooperation is very useful but often ignored by economists and politicians as it goes against some common sense rules.

The entire theory of probability is counter-intuitive and has plenty of traps in economics for those who use only common sense. One example is a Monty Hall paradox: once you made your decision with limited information, you will gain by changing your decision when you get more information. A common fallacy of "throwing good money after bad" stems from ignoring this principle. Literally everyone has been guilty of it.

• The biological examples are nice, but explaining them would greatly improve this answer. "throwing good money after bad" has nothing to do with Monty Hall. It is a coping mechanism for the cognitive dissonance arising from having made a bad decision. – Giskard Mar 18 '18 at 10:19

It is rational to defect in the Prisoner's Dilemma

Cooperation is socially optimal in the Prisoner's Dilemma, and in this economists and non-economists typically agree. However, the rational, i.e. the individually optimal choice, is to defect. From the game-theoretic point of view this is trivial, since the payoffs of the Prisoner's Dilemma game make defection a strictly dominant strategy. But it is notoriously difficult to get non-economists to agree to this reasoning. After all, if both prisoners cooperate, this makes both of them better off compared to bilateral defection, so isn't this proof enough? Ken Binmore’s (2007) Game theory: A very brief introduction contains several different lines of reasoning people have come up with to support cooperation as the individually optimal choice; all of them flawed.

The best answer to this question has to be pricing. It is agreed in economics that pricing reflects the supply and demand and it depends on these on a free market. Still, you'd often hear people how some product is too expensive because it is just some raw material plus some working hours.

Even worse, you sometimes see business owners billing their service or products by some amount of material and some hours of labour. For e.g. a carpenter who sells some closet for 1.2 cubic meter of wood at 800\$/cubic meter + 30 hours of work at 90\$ per hour, giving a grand total of nobody cares. Because if someone comes in and says I want that closet for 1 million euros and you're willing to give it for that money, that's PRECISELY what it's worth. Exactly in the same way as if someone comes in and wants to pay you 100 euros for the closet, you can't claim it's worth at least the material+labour, because nobody is willing to pay that money, so it's not worth that much, but PRECISELY 100 euros.

In a nutshell, it is a consensus in Economics that prices are driven by supply and demand on free market, but for some reason people very widely believe that it's common sense that prices are driven by production costs.

• uhm have you ever sat in a micro lecture? Because most of that is arguing that in equilibrium on a competitive market prices will be equal to costs. Because if it is above cost you will have entry and below cost firms will close. Which means you end up with prices at cost. And this makes markets efficient. If everyone who has higher utility from the product than it costs gets it, aka when the price equals cost, then you get a pareto optimal outcome. Effectively when considering a product, you consider if the value it provides is higher than the cost it causes (the price). – Felix B. Mar 14 '18 at 20:39
• Actually, the price is equal to the marginal cost of production in any efficent market. – Repmat Mar 15 '18 at 13:34
• @FelixB. and repmat Both of you made my point. People noticed that there is somehow a correlation between production costs and prices, and immediately assume that prices must be driven by production costs, which is something economists agree it is not true. It's as if you're watching a carriage with a horse and deduce that the carriage must push the horse further, since they're both going at the same speed. – Andrei Mar 15 '18 at 13:59
• @Andrei but you complain about "you'd often hear people how some product is too expensive because it is just some raw material plus some working hours" and if you agree, that in a competitive market the price should be marginal cost, then you also have to agree that if it is not at marginal costs, there is a market failure and the people are right to complain about the wrong prices. So it is a fine thumb rule to consider the costs. – Felix B. Mar 15 '18 at 18:57
• You are all literally making my point. You just don't get how the market works and you're also not reading the answer. Any price is simply an agreement between the buyer and the seller. None of you answered why wouldn't a transaction happen when the production costs and the price are far off, if the buyer and the seller agree. All of you suggest that that can't happen when it is happening everyday. See cpu prices, airplane first class seats, fashion, etc... I am actually glad that I have the lowest score on this answer, as it stands as proof that it's something easy, yet nobody gets it – Andrei Mar 17 '18 at 13:39