Economist Eugene Fama: 'Efficient markets is a hypothesis. It's not reality
ft.com
ft.com
The market (in any good) is an information processing system which tries to collate a colossal number of variables down to a single one: price. It has the same limitations as other information processing systems, as well as things you may be familiar with from control theory like frequency response.
It seems like we have an economy which is highly optimized at reducing price, but extremely primitive at making “quality” transparent enough that price can be properly assessed.
A case in point being Amazon marketplace where you can find very low priced products from completely opaque brands with almost zero signal on quality.
Such review systems are trivially gamed, with exceedingly high incentives to do so.
One of the underappreciated benefits of an established retailier with limited suppliers and a centralised buying operation, as with traditional brick-and-mortar establishments (Harrods, Costco, Macy's, etc.) is that those establishments can set their own minimum standards and select merchandise which is highly likely to provide high value to the end customer. The notion that "choice is good" falls flat in the reality that uninformed choice over a range of intentionally insufficient products is not good, and is in fact a form of gambling with low payoffs.
Some retailers target high-end products (e.g., Harrods), some mid- or low-market (e.g., Costco). But by restricting upstream vendors, tracking satisfaction, and having their own testing and development of products or offerings, as well as return/warrantee programmes, they can assure a reasonable rate of satisfaction.
I'm also finding retailers who sell manifest crap (o hai any Samsung home appliances) can be eliminated from my own list of acceptable outlets for future purchases.
Oh sure, absolutely, but note I said "easiest to implement". And there's that cost measure again, as if it's the only KPI. How do we measure the quality of the "quality" system? :)
(I think your question's a valid one, and a quite good one ... a high-quality question, if you will. I also Have Thoughts, but would prefer to see you expand your question before polluting the idea space with my own biases and opinions. Though if I may steer the conversation somewhat: audits. Of what and how would be informed to your response to my first question though.)
It also starts with what the goals of a rating system are, and how it might be subverted or attacked.
I see four key principles or assumptions:
1. The reported ratings for overall product performance should correspond with some reasonable informed-experts' ratings.
2. Issues with product support (returns, refunds, repairs, replacements, etc.) should correspond with identifiable events.
3. Solicitation of ratings, false positive ratings, or false negative ratings as forms of harassment or industrial subversion are all obvious (and extant) forms of abuse, as would be the ratings platform itself predicating positive reviews on, say, advertising purchases by the reviewed site, as has been alleged in numerous cases.
4. Systematic abuse will tend to be orchestrated by a given entity, and should exhibit some form of behavioural or networked anomalies.
For key products (high volume, high-price, and/or both), independent assessments should be performed on at least some sample of products and conformity with assessed ratings shown. Where expert opinions can be reasonably attained, they should generally take priority over crowdsourced reivews, especially where specific expertise and tangible distinction in results exists. This is guided by both the notion that you'd rather have a skilled pilot than a democratic representation of passengers flying an aeroplane (a notion dating back to Plato, though he somewhat unambitiously used the example of a ship at sea), and awareness that in some cases there isn't much consistency even amongst experts, as with wine tastings and ratings. Third parties (e.g., Consumers' Union, Underwriters Laboratories) might also be compensated for unbiased expert assessments.
Individual raters with high volumes, anomalous ratings, or patterns of coordination with other raters (e.g., similar ratings on the same products) should be flagged for review, and potentially voided from the system.
Any indications of suppliers soliciting ratings, providing kickbacks or rebates, or otherwise inducing positive ratings should be addressed severely. I'd be in favour of a quick and long-term minimum-rating consequence. This of course raises the risk of suppliers "joe-jobbing" competitors' offerings to create the appearance of manipulation (by the target). That activity should result in an even harsher consequence.
The overall rating system and activity should be constantly monitored for anomolous patterns. Highly-active individual accounts are probably the easiest to detect. Large groups of low-activity accounts with similar activity are probably the hardest to detect, but would be another key issue. Audits of such behaviours should be a constant part of a system, and the subject of regular transparency reports.
For instance, forced labor, pollution, etc. are factored into price. Or more specifically, the customers willingness to care about these things is factored into the price, on the buy side. And someone’s willingness and ability to do them (or hide them) is priced on the sell side.
Same with the customers willingness to gamble on quality (aka random no name branding), and a sellers willingness to deal with legal and reputational blowback.
People just don’t want to say it.
At the end of the day, as long as no one is going to shame them or punish them, most people will happily fund super polluting slave labor and gamble on quality for a cheap price.
Hell, as has been made clear over and over again, people will happily risk serious legal consequences and even murder people to supply/consume illegal goods that get them high, at a price. And that price is a lot lower than most people want to think about.
And this is where the EMH bears actual fruit - because the ultimate test is people’s willingness to part with their cash for a good. And both sides of the transaction are always trying to figure out ways to shift the line where they both meet in their favor.
It’s not about what they say, what they profess to believe, what is acceptable to admit, what is legal (per-se), etc. unless it actually matters.
Which is rarely as often as we’d all like to believe.
I don’t agree with this. Pollution primarily affects people who cannot afford to pay to not affected by pollution. The people who buy the most polluting products are not the people who are directly exposed to the most pollution. This isn’t as simple as “oh people just don’t feel bad enough to not buy cheap things”, the people who cannot afford to avoid the consequences of societally bad decisions are suffering way beyond their consumption! The poor live most exposed to effects of climate change (fires, floods, etc) because they cannot afford to avoid it, not because they somehow calculated that they don’t feel bad enough about it. The wealthy are not forced to have their house flooded in accordance to the number of cars they own, for example.
That's not to say we wouldn't spend $5.40 for entirely ethical production; We are not granted that choice. In a wildly unregulated market with little oversight, establishing the chain to demonstrate ethical viability is expensive, and only people who have a lot of income and spare time can afford to even be worried about this.
Right now, ethical production in many areas of world trade is dominated by a fringe marketting strategy to price-segregate customers, because nobody has even tried to regulate a better outcome.
Expecting people to spend many hours studying every single thing they consume for signs of malfeasance, especially when most of the data they would require is a trade secret or things which manufacturers are actively incentivized to lie about, is crazy. This information black hole provides a second limit to effective pricing.
* American double-pane sash windows on hardware store shelf: $100
* European triple-pane tilt turn windows, on hardware store shelf: $200
* American triple-pane tilt turn windows, 'call for quote': $900
...
* No Salt Added Snacks in the health food section for 10% of people: $4
* Lots of Salt Added Snacks in the main section for 30% of people: $2
* The 60% of people who would prefer a moderate amount of salt for $2: Not a part of the Nash Equilibrium
There are all sorts of areas where the market offers suboptimal choices because consumer pricing, availability, product segmentation, and market segmentation is a game that is somewhat orthogonal to the range of consumers' ideal outcomes. Sometimes, regulation significantly improves those outcomes in a way that prices cannot. Banning slave labor is one example where it would be a significant benefit for a trivial cost, but the market absolutely isn't structured to pay that cost using higher prices.
Externalities not paid by consumer or manufacturer is one category of those benefits to examine, but there are many more.
There are, for many clear domestic socio-political reasons (including highly embedded religious and idealogical ones), many countries which are never going to be effectively limited in their untrained labor force, or that have the ability/interest to enforce good labor practices.
And there are too many ‘buying’ nations to effectively enforce a Monopsony either.
Prices are a reduction of information in that since decisions that lead to price being set are past that information gets reduced.
Or to put it another way for an investor speculator they have to go to the chain that goes into producing that product to find the signals to use to short or buy a stock...not the price of the good itself..
If the buyer cannot estimate the quality of the product before buying, the sellers will reduce quality or be forced to by competition. Buyers lose trust and the market can collapse.
The reason why people don't pay for online stuff is mostly because we're used to get free stuff online + the fact that online payment remains cumbersome (even moreso since the businesses have all incentives to pish you to a regular payment model instead of a one-shot one). It has nothing to do with information asymmetry.
And that a paper that reads like a parables like this was deemed Nobel-prize worthy tells us more about economics as an academic field than about the paper itself.
You can't tell from the output price which selection was made.
> SSRN provides 1,453,207 preprints and research papers from 1,847,303 researchers in over 65 disciplines.
It's basically like Arxiv, but focused on different fields. The articles may be preprints of peer reviewed articles published in leading journals, or they may be mad ravings uploaded by a crank, or anything in between. Also see wikipedia's article on SSRN[1].
From SSRN's page on Algorithmic Finance[2], the supposed journal (reachable by following the link under the article title):
> The journal archives all papers on SSRN
Note that the journal managing editor and contact is the author of the paper, the contact info appears to be his personal contact info, and all links on the page are dead.
[1]: https://en.wikipedia.org/wiki/Social_Science_Research_Networ...
[2]: https://papers.ssrn.com/sol3/PIP_Journal.cfm?pip_jrnl=167527...
> Note that the "journal" editor is the author of the paper.
Interesting, I didn't notice that. The journal does have other editors too, see the first page of the PDF.
Efficiency can also include non-tangible value. A lot of people want to own Tesla shares because it makes them feel cool or techie, so it trades at a huge premium over the fundamentals. But it doesn't change the fact that if there is an expected increase or decrease in company performance, the going share price will still adjust amazingly quickly.
So "efficiency" is really in the eye of the beholder. One person can look at a circle from 10 feet away and say, "look a perfect circle!" and the other can say "there are no perfect circles" and the second person can be technically correct, but the first person's observation may be perfectly valid.
This isn't too much of a thing for stocks (most likely because they aren't production inputs), but that's a small part of price signals. In general, the most important prices do not adjust amazingly quickly, because our economy is far too slow and unwieldy and unable to cope with price uncertainty.
Efficient markets hypothesis was written specifically with stock markets and exchanges in mind, and it's what the linked article mostly discussed.
Stock markets are ultimately downstream of production, it is impossible to divorce the efficiency of commodity markets from the efficiency of capital markets. Rigidity is contagious.
Firms with sticky prices are wildly considered to incur costs and more negative responses to various macroeconomic conditions, and those costs reduce the valuation of equities, so the inefficiency in production is reflected in equity prices. If prices were not rigid we would see different equity prices.
You can try to fix this by looking at every transaction level in isolation and baking in the bias of the next level, but then you're no longer making a statement about the economy and the efficiency of price signals.
Instead, you're making an essentially unverifiable statement about the profitability of a limited set of strategies restricted to long/short positions in equities only (which on average is forced to be exactly equal to the market price since every transaction needs a counterparty):
Even if there is a hidden variable that makes some strategies better, it will most likely be impossible to pick out which strategies are better because of luck and which are better for a reason if you are modelling the null hypothesis as some sort of random walk with long tail event, the hidden variable will most likely be lost in the noise.
You can also try testing it by looking at prices directly, in which case Roll's critique holds.
In short, as a broader theory to apply to economic relations and economic calculation, the EMH doesn't hold. As a narrow theory of the profitability of certain trading strategies, it's unverifiable.
The problem with most of economic science is that latter part. Where most other sciences try to decouple themselves from their subject or at least aknowledge the limitations of their field, many economists seem to have a much more loose relationship with their field. E.g. if a hypothesis failed to explain so many real world phenomena as many popular economic ideas, any natural scientist would throw their hands in the air and admit they had no idea what was actually going on. Yet my feeling isn't that economists are like: "We have that part here figured out and that part over there is work in progress" but more of an airquotes-science, like Lacanism or Freudianism, where what you say depends on whose school of thought you subscribe to.
Granted if one imagined how a truly serious about describing-what-is-economic-science would have to work like, that would be a gargantuan endeavour as it not only encompasses mathematics and complex systems, but also fields like psychology, sociology and geography. So I am the last to blame any serious economist for not being able to explaining everything. What I can however blame them for is when they defend a hypothesis that has been falsified by reality countless times by moving the goal posts by saying the conditions have to be just right for their hypothesis to make sense. Yeah sure.
Regardless of how much efficiency there is or is not, it's still a good theoretical framework for understanding markets. No one assumes the cost of a wheat future is a completely made up number so obviously there is some efficiency.
I have nothing against useful tools or models. But all too often these get confused with reality in economics — with potentially world-changing consequences. Granted this wasn't the point of the article, it was more like a tangential thought.
The old guard still runs the show, of course, and the field will advance one death at a time.
Many people would agree that efficiency is good while exploitation is bad, but we shouldn't let our feelings toward each idea cloud our understanding of the relationship between them. It's very possible for a concrete situation to be both good and bad at the same time, depending on which part of it we look at.
Amazing that 100 years later we’re still unable to grasp the nuance of the efficiency idea.
We settled this one forever ago, but somehow armchair internet commenters still think they can zing the entire field of economics by saying “lol you guys still think people are efficient robots.” Cool 1940s insult, bro.
Yes, obviously markets aren’t perfectly efficient and nobody believes they are. But they trend toward efficiency over time. Ignore this at your own peril.
In fact, one of the main errors in Karl Marx’s theories from the mid-1800s was the belief that markets were perfectly efficient. He thought profits in a market economy would inevitably fall to 0.
It turns out humans constantly want new and different stuff, in unpredictable fashion, and therefore this never happens.
Another one being the Intrinsic Theory of Value which he borrowed from Adam Smith. Marx allegedly postponed the publication of the 2nd volume of his Das Capital, when he learned about Marginalism, as it would require a complete revision of his theory, but died before he could do it. Von Mises writes that if that's true, Marx was way smarter than his followers.
Well that's generally the case, I'd sort of be surprised if he weren't.
I suppose that's efficient in its own way, if you're one of the relatively few oligarchs and wannabes who benefit from it.
To anyone rational and politically literate, these and other expedient mainstream superstitions are obvious nonsense.
What does efficiency mean here? And can you support this claim?
So it's not quite the same as code inefficiency, unless an engineer could reap the reward for the fixing of said inefficiency.
To be fair, the "Bug Free Code Hypothesis" is kind of true. Because there are lots of incentives to eliminate bugs, people do put a lot of time and effort into doing so – yielding modern miracles such as OS kernels with 30 million lines of code that almost always work as intended.
But while software is bug free to a perhaps surprising extent, it would be foolish to plan a software project on the assumption that there will not be any bugs to fix, or that a small number of bugs cannot have a big impact. Similarly, it seems unwise to assume the efficiency of markets in economic planning (though I don't know to what extent economists actually do this).
Most of those programming management memes come from big companies with giant teams that don't live or die on output, many of them are shielded from that sort of thing because of cash cows like Microsoft/Google or corporate megadeals like IBM/Oracle. So there's never any shake ups until it's long been obvious to the customers.
I'm also not referencing any programming management memes.
I think maybe you're seeing a cynicism in my comment that's not there. I don't think that bugs exists because of dumb managers. I think they exist because it's very difficult to write code without any bugs, even when there are very strong incentives to do so. That's partly just because it's inherently difficult, and partly because there are also other incentives pulling in different directions.
More money has been earned by removing the GameStop inefficiencies than by creating them. GameStop shot up in value driven by WallStreetBets, but those people mostly lost money by creating the inefficiencies.
The main inefficiency in GameStop is creating information and will to buy that runs completely orthogonal to the actual valuation of the stock. Like snake charmers, Roaring Kitty etc use memes and bots online to drum "bagholders" into a fervor, thinking they are righteously sticking it to some nebulous man by buying a stock that is worthless.
that's called a pump and dump, which is already illegal. Because the fraud depends on manufactured (mis)information.
The trouble is that, even if this is true, there is no theoretical upper bound on how long it can take. If "markets trend to efficiency" means "don't worry, wait a few years and this inefficient market will be efficient again", then that's one thing. If it could take a couple of hundred years, that's quite another. As Keynes famously put it:
> In the long run, we're all dead. Economists set themselves too easy, too useless a task, if in tempestuous seasons they can only tell us, that when the storm is long past, the ocean is flat again.
There's an interesting analogy with non-Turing-complete programming languages that guarantee program termination. In the real world, a guarantee that a program will terminate eventually is often not particularly useful. What we really want is a guarantee that it will terminate within some reasonable period of time.
EMH is the closest finance has to a “theory of everything”, and won Fama the Nobel Prize for Economics in 2013. But it remains as controversial today as it did when Fama first proposed it half a century ago.
(From TFA.)
The fact that the hypothesis's own formulator and presumably chief cheerleader now has his doubts is worthy of comment.
Fama has always made the distinction that it's a hypothesis, not reality.
e.g. 'I start my class every year by saying, “These are models. And the reason we call them models is that they’re not 100 percent true. If they were, we would call them reality, not models'. https://www.minneapolisfed.org/article/2007/interview-with-e...
So I agree with the GP, "I don’t know why Eugene Fama saying this in 2024 is considered news."
Which makes litigating of the title tedious, and failing to fullfil HN's objective of satisfying intellectual curiosity. Dang explains this (in a somewhat different context, but otherwise relevantly) here:
His key EMH paper [1] notes that a generic formulation of the EMH, that security prices at any time "fully reflect" all available information is untestable: you need narrow down the information set and the pathways in which they are reflected in prices. The paper goes over various technical definitions of "efficiency" and concludes from reviewing the empirical literature (in 1970) that no evidence against certain formulations (weak, semi-strong) of the EMH has been found/published.
I don't think it's fair to characterize Fama as "chief cheerleader" of the EMH, but the papers are there for the readers to judge for themselves.
[1] https://www.jstor.org/stable/2325486, http://efinance.org.cn/cn/fm/Efficient%20Capital%20Markets%2...
Economic theory is kind of bullshit, they're building predictive models of enormous complex systems and trying to ascribe meaning, the models can be "ok" but the meaning the economists give is personally biased hand waiving about 100% of the time.
On the contrary. It is a religious chant on the Right, that Free Markets are efficient, and The Free-er, the More Efficient, and will solve all of our problems.
--> "All hail our god the free market, efficient in its repose, elegant in its demonstrations, arbiter of logic through the efficient drive of profits, holy in its dispensation of moral judgement, let those that are poor be scorned for the market has judged them unworthy of its riches".
Though, I can agree, maybe it is, most people all around, only have a simplified view.
And yet all the currently popular theories (and narratives) of economics are founded on that assumption.
The F-twist is really someting to behold, especially as the theories don't even work.
Do you actually believe everyone in the entire field of economics believes people are robots? Or is it just convenient for you to hold that belief since it allows you dismiss an entire field of human endeavor in one sentence?
They don't, but a lot of them work (and preach) as if they are.
This is not perculiar to economics, but rather a general practice of scientific programmes in the Lakatosian sense. But few fields cause such harm with and are so politically motivated to protect their core assumptions.
Newtonian physics is 'popular' because it is easy to grasp with a high school education.
Likewise, the theories of the early David Ricardo era are easy to understand, maybe popular, but not current with academics or practitioners.
Because there are still significant numbers of people, many of whom are in positions of power, who think that increases in market efficiency solve all problems in a society.
When you have people thinking that shortages of things like "food", "medicine", "shelter", and "water" can be solved by simply adjusting prices until some people can no longer participate in the market, you don't have a hypothesis, you have a cargo cult, and a psychopathic one at that.
The market is rife with information asymmetry that seems to be increasing, and for it to trend toward efficiency, we would need that to be trending downward.
Markets become more efficient as a result of entities who make them so. It is possible to make a very good living by being one of these entities.
In fact it's kind of the opposite of efficiency, because when the efficient market hypothesis is true, then it means that the market makers are losing money compared to the passive investors, and that's very inefficient: it gives nobody an incentive to perform the work of capital allocation and instead incentivize everyone to be freeloading on other people's work.
Its weird Zuckerberg has accumulated all this data about social dynamics and no theories have emerged yet.
To start a business of any kind is to hypothesize the existence of a market inefficiency, something the market is not currently doing that would make it more efficient. All business is in some sense arbitrage between state A, the market as it currently exists, and state B, the market with your business in it. A business succeeds if its hypothesis is correct and if it executes well.
Also there's this: https://arxiv.org/abs/1002.2284
In fact if the EMH was true, then all hope for financing innovative companies would come from delusional investors making the sub-optimal choice of investing there. How “efficient”.
Why? Then how would be all your stealth companies supposed to raise enough money in the first place? Or attract customers?
In practice most companies are publicly known as early as series A at least and remaining unnoticed for longer requires lots of effort that would definitely hamper the growth of such companies.
And the market definitely takes rising competitors, even private, into account when assessing the financial prospect of existing public companies.
That definition of EMH seems to be what's causing this idea that innovation in new companies is impossible, rather than the general idea that asset prices reflect available information. Some information is not available, and that won't be reflected in prices. That's all.
emh doesn't mean everyone knows everything, it means the price accurately reflects available information. if an innovation was believed to be world changing, it would be worth a lot, if the belief changes, the price changes.
The economy is a faith-based construct. Prices aren't set by "information", but by faith in value - which is one of the easiest things in the world to lie about.
In EMH there is cooked in some notion of overall largely rational beliefs, and yet the shares trade hands from someone selling to someone buying - both of which feel they got a good deal worth making money from.
Efficient-market theory doesn't apply to startups because they aren't listed on the stock exchange. The theory doesn't apply to small number of investors like VCs. It definitely doesn't apply to the business of the startup, or to the reaction of the market to the startup. It does apply when the startup goes public.
The efficent-market hypothesis is not general supposition that for example all markets are efficient by all definitions, nor that free-market capitalism is efficient. It's a specific hypothsis the prices of financial assets like stocks represent the true expected value[2] of the assets, including when accounting for future retuns. The supposed mechanism driving the property is that finacial asset markets are effectively servo systems[3]; traders trade based on any measureable information and act as feedback which drives prices to the expected values, so any other variables which may affect prices are therefore unmeasurable.
Personally, it seems clear to me the hypothesis is literally false. Someone has to have the most actionable information or be the best at measuring. On the other hand, it seems clear that it's practically true for almost all people. The chance that any given person can measure an asset's true value the best is practically negligible. I certainly don't believe I can beat the market.
[1]: https://en.wikipedia.org/wiki/Efficient-market_hypothesis
https://www.nobelprize.org/prizes/economic-sciences/2013/sum...
I was with you until this step. That markets are not close to efficient (in the sense of EMH) seems right to me. BUT it does not follow that because of that anyone should be able to beat the market.
There is a huge gap between EMH being far from reality (or simply not close) - and that gap being exploitable. (Or more generally, non-EMH doesn't say much as to whether or how things should be exploitable.)
For one thing we humans are humans, report to humans, get our news from humans, share this market with humans, etc, and the ones of us who pay attention know how hard it is to account for that.
That they are efficient in communicating what a company is worth on fundamental financial level is laughably incorrect and not even worthy of "in best case scenario" theorizing.
What they are efficient at is capturing literally EVERY possible human motivation to buy, sell, or ignore, and distilling it into a single number.
It’s a pretty good tool for analyzing markets and understanding what’s happening. Like anything, it’s impossible to create an abstraction that’s 100% correct or works in all circumstances when discussing a massive dynamic system that is the real world and life.
The study of the spread of memes is something that's the subject of research currently.
- human resources
- headcount
- talent equity
- talent pool
- cogs in the machine
And countless more. We’re already priced as assets. Openly.
On the other hand, nobody sensible believes they are completely inefficient, or trivially beatable. If you want evidence of this, feel free to go out and beat the market with long plays on the alpha you imagine you have.
I think it's usually helpful, when trying to understand the world, to think, "markets are pretty efficient, do they agree with me? If they don't, can I think of a real reason why this may be a case where they're wrong, or am I just engaging in wishful thinking?"
I think it is not usually helpful, when trying to understand the world, to be really mad about the EMH.
If the market is chaotic, you can't beat it just because you're smart.
1. Just buy when the price is low and random variation will move the price higher eventually (or short when the price is high).
2. Ignore the market except to buy when a company is underpriced, then just have the company declare dividends or whatever and directly eat the profit that the market is stipulated to have underpriced.
A very small investor in a random market might have difficulty with either of these strategies, but a reasonably well-capitalized investor would not.
If you want to think up some kind of complicated model of difficult-to-take-advantage-of company pricing that is hard to exploit, really ask yourself whether that model is grounded in anything other than, "I'm mad about the EMH."
1. How do you know when the price is high or low? The market can remain irrational for a very long time (longer than you can stay solvent).
2. You can easily lose money shorting when you're right. People do it all the time.
You can't blow up.
You're not the House at the casino.
This is what I mean when I say that it's usually not helpful to try to make claims about the world based on, "I'm really mad about the EMH." You end up making silly claims.
Geometric Brownian motion can't take negative values either, so I don't know what you think this proves. While it's true that stock prices don't literally follow geometric Brownian motions (+drift), you can't tell the difference from stock graphs. (You can tell the difference with statistical tests if you know what to look for, e.g. volatility clustering, but humans can't tell by eye, and naive classification ML models will not perform well.)
I will note that "stocks are priced via geometric browning motion" would satisfy the EMH. And that the underlying growth trend of GBM, in a real world, is, like, "the stock market actually values real things about the company that you can find out, but which other people have already found out."
- "The law of supply and demand"
- "Trickle down economics"
or, my favorite:
- "The invisible hand of the market"