I think the investment floor argument here is actually understated.
Successful investors rapidly find themselves in a world where the vast majority of stocks are too small to buy. Try putting on a multi-million dollar position on for a stock that trades $10k/day on average.
Even index funds struggle with this. There are "small cap[italization]" index funds that have a median firm size of well over $1.6 Billion dollars larger than the vast majority of stocks.
This is reflected also in research showing that fund managers do not exploit the small cap outperformance anomaly nor the fact that other anomalies (like value) are far more powerful in small cap stocks.
You see this reflected also in the reversion to the mean of fund performance as they get larger.
Other disadvantages of the smart money are documented in the paper "the limits of arbitrage" https://onlinelibrary.wiley.com/doi/full/10.1111/j.1540-6261.1997.tb03807.x
One issue is that fund managers, who control most of the investment funds, live and die by short term performance and cannot implement any strategy that underperforms the relevant index by 6 months or so. This is why fund managers typically do their mean variance optimization on tracking error, not on total returns.
They are in effect only as smart and well informed as their investors. See the 2018 paper "What Do Mutual Fund Investors Really Care About?" for some information on this.
The other main issue is that mispricings can get far worse before they are resolved. This makes it very hard to bet against them. Several people who bet against the toxic waste (subprime) mortgage trusts before 2008 lost a lot of money and were forced out of the trade. See the paper "Toil and Trouble, Don’t Get Burned Shorting Bubbles" by Aaron Brown et al
> After analyzing the data and speaking with traders who actively researched
subprime mortgages we highlight latent risks that were associated with shorting subprime
mortgages. These latent risks help explain why many smart, informed traders decided not
to short subprime mortgages ...
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3782759
Having said that it is actually very hard to beat indexing. People should assume this is a serious and difficult endevour requiring lots of hard work on learning the field and on their own psychology.
Some factors which I think are both important and missing from your model:
I agree, these are all important and missing.
The concept of this model isn't to give license to any trade meeting these criteria. Instead, it's to show you a category of trades that might be ignored by hedge funds and HFTs not, a priori, because they are bad investments. So the idea would be that within this space, you'd then look for trades with an attractive risk/reward/tax/fee profile.
I do agree that opportunity costs might be the clincher. It might be that no matter how much you earn an hour (risk adjusted), the simple fact that you can make that much through this form of work is strong evidence you could have made more in another line of work.
It points to a model of the world that goes something like this:
Humanity has a giant pot of slack, known as "funding," which it doles out liberally to people who've got a reasonable chance of providing value to shareholders. This tends to generate even more wealth for those value-generating businesspeople, who then have nothing to do with that money but put it back into the pot.
Investing, then, is only for two kinds of people:
So maybe we need a change of quote. Instead of "if you're so smart, why aren't you rich," it's "why are you investing, if you're not rich, retired or dumb?"
There are some commodities plays in which large scale investments perform almost as entirely different commodities than small scale investments.
Personally I saw this in late Feb/early Mar of last year. I had bought some physical silver bars, and during the 'flash crash' saw that the price of spot silver (COMEX 5k oz contracts I believe) had dipped to $12. But at the same time every online retail store selling 1oz silver bars was sold out, and I was able to sell any inventory I wanted for $35 an oz online.
Now ignoring weirdness in the precious metals market and COMEX manipulation conspiracy stuff, I think this was due to exogenous market conditions forcing a mass sell off in everything including silver contracts. And this big discrepancy in 'retail price' (price of a 1oz bar) vs 'bulk price' (price of a 5k oz COMEX contract/5k) not only corrected but in fact led to a pretty big silver bull run all the way up to ~$28. But more importantly the point is in reference to your post, if you were a big institutional investor with a $1B fund you probably couldn't afford to pay the premium to buy 1oz bars originally, but you also would have gotten wiped out of your COMEX position if you even had a little bit of leverage trying to get exposure to silver. On the other hand if you were a retail guy like me just buying 1oz bars online, you made a bunch of profit.
Hope this helps.
The type of alpha trading available to retail investors tend to be slow-converging, noisy inefficiencies. Things like statistical arbitrage (pairs trading/baskets), or style-factor trades (illiquidity, momentum, carry, etc.), which persist for economically plausible reasons. You have to be willing to very systematically grind out small edges over a long time, because often about half of the trades are losers. You also need applied data science to find these and notice when they dry up. "Gut feels" are all but useless in the face of this much noise.
Big investors, on the other hand, prefer to exploit fast-converging supply-and-demand imbalances. High-frequency trading, direct arbitrage between exchanges, dispersion trades, etc. You need a lot of capital and staff, but the trades are much more reliable, in the sense that if the edge disappears it's easy to notice that quickly. Why would they bother with the slow trades when they're busy exploiting the fast ones?
I know of an investment that fits all of these criteria. Retail scalping. Due to the fact that you can usually return the items you bought if you fail to sell them for above your cost, the risks of a loss is low. It's small scale - there might be '10' of the desired commodity in an online store showing up in a day, and your bots could snag just 1. The ROI can trivially be 20-50% in a week as you 'flip' the item for a markup.
It is generally considered to be despicable behavior but is also currently legal.
Downside risks : sometimes retail scalpers have gotten ejected from online marketplaces for hoarding essential goods. For example early in the pandemic there are some who hoarded hand sanitizer and then they were banned from selling them online.
I don't practice it myself so I don't know all of the risks, but it seems to fit your request.
Yes, this is a good example. I think a real gung-ho capitalist would say that these people are doing a public service, and are just engaging in the same sorts of price discrimination that conventional businesses do all the time.
In a way, a bookstore is doing the same thing. An author's book is sold as a preorder for lower than the later retail rate. The bookstore purchases a stockpile of copies at a low price. Then, when the release date rolls around, they charge a higher rate than what they paid. If their customers had only known about the preorder in advance, they too could have snagged it at a lower price. Now, they're forced to deal with the book seller.
I also know somebody who runs a business where she buys cheap clothes on Ebay, models them, and then resells them on Instagram for a steep markup. That's retail scalping - with a small value-added service, to be sure - but it doesn't feel repugnant to anybody who hears about it. It's just a step above dumpster-diving. You're looking through a bunch of things people are giving away for a low (or zero) price, picking out the few that are actually nice enough to warrant a higher price, and then charging a premium for your service.
It would, of course, be nice to find such opportunities with the additional requirement that they're not repugnant. But in principle, yes, as long as the deal isn't fraudulent, anything goes. The question is whether there's money to be made, not whether there's nice-looking money to be made.
But for someone who's interested in the latter question, they just have a harder problem to solve. The same thoughts above might help them to find solutions.
How does price discrimination serve the public? I got the impression it's the sort of drawback of markets that gets regulated away when it gets pronounced enough, like network effects.
I think there's a good chance my answer might change if I thought about this more. But the underlying thought is that price discrimination gives consumers more choices.
Coupons are a real-world example of PD. Without them, consumers have two choices:
With coupons, consumers have a third choice:
3. Consume at a lower price, in exchange for the fuss of keeping track of coupons.
More choice is good. Price discrimination allows people who'd otherwise be priced out of consumption to get more of what they want.
The sourcing-cheap-clothes-on-eBay method doesn't sound like 'small value-add' in that evidently people are prepared to pay quite a lot extra for her to do this. I.e. to save them the trouble of trawling through pages of crap on eBay; but also presumably saving them the risk that stuff sold on eBay (with crappy photos etc.) is often crummy, whereas stuff properly modelled (with nice photos) is more likely to be good.
Right. It's arguably not morally worse than various HFT and dark pool and other fintech moneymaking tricks though. All these involve buying a mispriced commodity (even by a fraction of a cent) and reselling it for it's true (market value). And the buying/selling opportunities are unavailable to most people, in the same way you can't retail scalp effectively unless you are using a computer program to do it.
My point is the 'less morally repugnant but also still as profitable' hurdle isn't an easy one to clear since it's not that morally repugnant.
EDIT:
I'm now realizing that I probably misinterpreted Gerald. I think he means people who set up an e-commerce site selling, say, a dog toy, at a higher price than you'd find on other e-commerce websites. If the customer buys it, the retail scalper orders the toy through the lower-priced page and has it shipped to the customer. If the customer wants to return it, they have the return sent to the original seller, who eats the cost of the return.
Effectively, the retail scalper is exploiting (if not stealing outright) the "free returns" offered by the original seller. This isn't just about stockpiling, or putting in work to find underpriced goods and sell them for what they're worth.
Instead, it's about exploiting people who are bad at comparison shopping, and market norms that enable things like free return policies to exist. It raises prices for the consumer (who might naturally expect that the vendor they're buying from isn't a retail scalper), and imposes an hidden risk of extra costs on the seller. Consumers who realize they've overpaid for a product might return the version they bought through the retail scalper to the normal e-commerce vendor, and then turn around and buy it right back again from that vendor through the normal e-commerce site.
The best defense of it that I can make is that retail scalpers are adding a sort of "free advertising" for the product. Somehow, they're identifying consumers who weren't finding the cheaper prices on the normal e-commerce site, and convincing them to buy the product when they otherwise wouldn't have. But I'm not compelled by that argument.
For some subset of retail scalpers, they can be seen as taking on the risk of engaging in low-trust e-commerce; I can certainly beat the prices I pay for many of the goods I purchase, but doing so greatly increases my risk profile to, for example, somebody stealing my identity / credit card information.
For others, they provide a kind of availability arbitrage, by providing access to the same goods across multiple websites; I can pick the e-commerce portal I want to use, and trust that retail scalpers will ensure the goods I want to purchase will make it available there.
And for yet others, they are backing up purchases with their own reputation - pretty much every large retail company is reselling goods they purchased from companies in China which a buyer would be dubious about purchasing from; Etsy is full of resellers of Chinese goods in this fashion.
And on the subject of Etsy, there's also the dubious service of X-laundering; one example being ethical laundering, where a reseller sells goods from companies that people wouldn't purchase from themselves for ethical reasons; another example being status-good laundering, where a reseller sells goods that people purchase with the plausible misconception that the seller is handcrafting them. In all of these cases, the reseller is performing a service with regard to the original seller, and in some of these cases, the reseller may be performing a service for the end-consumer as well.
There's a wide variety of defenses for the behavior, basically.
I think this section makes two arguments:
I think both points are ~wrong / unimportant for EMH.
I think the argument (as you've written it) doesn't really make a huge amount of sense. (Saying that each person needs to find $x/per hour etc). Whilst not all trades scale, lots do, so finding a $10 opportunity is not necessarily easier or harder for a large fund than a $1mm opportunity. Some other advantages of scale which you're glossing over:
meaning that they'd need to be 20x as efficient as you in identifying them.
Consider the HFT - they are trying to place trades to earn pennies and they are much more than 20x as efficient as you at identifying those opportunities. (I don't think HFTs are particularly interesting to consider when thinking about "competition" in an EMH sense, but I think it's a good illustration about how much more efficient funds can be than you).
To be sure, a small investor still must compete against other smart, small investors for whatever edge there is to be found. But now we're basing the EMH, with respect to these small-time and strange investments, on the intelligence of people more or less like you and me.
I don't think this is quite the right interpretation. Can you describe to me a concrete trade which looks like: 20% return on $1000. (All the ones I can think of tend to be just as amenable to professionals). The other issue of playing in the "micro-investment" pool, is typically liquidity is much lower, so costs are higher.
EDIT: To be clear viz-a-vis your title "Are there opportunities for small investors unavailable to big ones?" I think the answer is "Yes, but there are lots more small investors"
Thanks for your thorough response. I hope you'll indulge me a bit more as I reflect on your argument.
Whilst each fund is structured differently, typically individual portfolio managers have a lot of autonomy within their mandate. If they change their mind about a position they can move very quickly. (Ignoring the size issue from the next paragraph). Bureaucracy can be a factor in some funds, but that's typically not what limits people.
"Inertia" is probably a poor choice of words. I really mean "investment in a type of investment," but that's clumsy-sounding. To be specific, I mean that I expect that hedge funds wind up with human and material forms of knowledge that's specially invested in a particular company or type of market.
In concrete terms, they employ people who are, say, experts in the Walmart corporation, and the tools (well beyond what's accessible to the average individual investor, such as those spy satellites looking at the parking lot) to evaluate it. That "Walmart specialist" isn't necessarily going to turn around tomorrow and start focusing on, say, Bitcoin. They're going to focus on continuing to find alpha in the area they understand best, and perhaps find logical adjacent types of investments if they're going to expand beyond it.
Extrapolate across the employees in the whole fund, and that amounts to something a bit like "inertia." Another way of putting it is that smaller investors face lower opportunity costs by exploring novel investment types.
Whilst not all trades scale, lots do, so finding a $10 opportunity is not necessarily easier or harder for a large fund than a $1mm opportunity.
This doesn't quite make sense to me, taken alone. Any $1mm opportunity is also a $10 opportunity, but not all $10 opportunities are $1mm opportunities. There must be more, say, $1,000 opportunities. Hard to give a firm reason to explain what the ratio would be.
Some other advantages of scale which you're glossing over:
This also doesn't quite fit the argument I'm making, though I accept all these points are true. The more hedge funds are pursuing investments that are unavailable to small investors, the more this works in favor of small investors. It means they aren't competing with hedge funds, because the hedge funds are pursuing other opportunities.
By the same token, I've been teaching piano lessons for the last 10 years, and I don't have to compete with Lang Lang for my business. He's pursuing other opportunities. If he showed up in my neighborhood and offered to teach lessons to the neighborhood kids, he'd have no trouble finding as much business as he desired. But because he has bigger fish to fry, I get to fry the little fish (so to speak).
In a way, the idea that hedge fund efficiency consumes all opportunities is like the lump of labor fallacy. Providing liquidity is a job, like any other. In general, getting more efficient at your job doesn't allow you to do all the work, leaving none for others. Instead, it leads you to specialize, leaving other opportunities available to others.
Consider the HFT - they are trying to place trades to earn pennies and they are much more than 20x as efficient as you at identifying those opportunities.
Agreed. I think that to be a successful small investor, you'd also need to find a niche that HFTs can't fill. I don't really know how they work, but I imagine they do things like technical analysis, sentiment analysis, and reacting to news reports automatically.
Humans might differentiate themselves from such algorithms by focusing on an approach to investing that considers it more along the lines of "superforecasting"-style questions. Figuring out how to ask and answer questions with good judgment, and determine how the answers bear on the market, is something that an algorithm is not cut out for.
I don't think this is quite the right interpretation. Can you describe to me a concrete trade which looks like: 20% return on $1000.
No (I'm assuming you're talking about 20% gains on a low-volume stock). But I have modest confidence that these opportunities might exist, and that I'd discover historical examples if I took the time to look. Right now, I'm trying to focus more on "how the EMH could be wrong (enough) under not-too-unreasonable assumptions to justify doing more research to verify or falsify the model," rather than making an argument that the model is valid or that the EMH is right or wrong. As I said, this is highly speculative, and I am inexperienced but trying to learn.
EDIT: To be clear viz-a-vis your title "Are there opportunities for small investors unavailable to big ones?" I think the answer is "Yes, but there are lots more small investors"
And this is my key point. Within this class of opportunity, you're not competing with hedge funds. You're competing with other small investors. One thing I really don't understand is what fraction of small investor wealth is tied up in things like index funds and managed retirement funds. Of small investor wealth, what fraction is people actively trying to play the small-cap low-volume weird-investment stock market? What level of sophistication does their thinking rise to?
I truly have no idea.
Extrapolate across the employees in the whole fund, and that amounts to something a bit like "inertia." Another way of putting it is that smaller investors face lower opportunity costs by exploring novel investment types.
Right, but then my point is individual funds don't matter. What matters is the ecosystem. Are there funds in your area? Another way of of putting your argument is: "Individual investors have an advantage because they can scan across all investments" but my counter to this is:
This doesn't quite make sense to me, taken alone. Any $1mm opportunity is also a $10 opportunity, but not all $10 opportunities are $1mm opportunities. There must be more, say, $1,000 opportunities. Hard to give a firm reason to explain what the ratio would be.
Lots of investments are $1mm investments and aren't $10 investments. (Try and see if you can get a $10 allocation in an IPO / VC / PE type investment).
Humans might differentiate themselves from such algorithms by focusing on an approach to investing that considers it more along the lines of "superforecasting"-style questions. Figuring out how to ask and answer questions with good judgment, and determine how the answers bear on the market, is something that an algorithm is not cut out for.
And you think hedge funds aren't doing this? I'm not sure exactly what edge you're ascribing to small investors here.
No (I'm assuming you're talking about 20% gains on a low-volume stock). But I have modest confidence that these opportunities might exist, and that I'd discover historical examples if I took the time to look. Right now, I'm trying to focus more on "how the EMH could be wrong (enough) under not-too-unreasonable assumptions to justify doing more research to verify or falsify the model," rather than making an argument that the model is valid or that the EMH is right or wrong. As I said, this is highly speculative, and I am inexperienced but trying to learn.
I like the way of framing EMH as "Is EMH true for you?" and there may or may not be reasons to believe it's true for you. Personally I have rather dim views of people who claim that EMH is not true in some generality. If you want to convince me EMH isn't true for you, develop a track record.
(Also to be clear a 20% gain on even the lowest volume stock wouldn't count for me - very low volume stocks would still allow investors to invest many times more than $1000...)
And this is my key point. Within this class of opportunity, you're not competing with hedge funds. You're competing with other small investors. One thing I really don't understand is what fraction of small investor wealth is tied up in things like index funds and managed retirement funds. Of small investor wealth, what fraction is people actively trying to play the small-cap low-volume weird-investment stock market? What level of sophistication does their thinking rise to?
I would recommend you spend some time on [insert any online investing forum]. There are loads of people doing it. I'm not sure the "fraction" is particularly meaningful. One thing which I think is mechanically true is more of the investments in small caps are active (since fewer indices invest in small caps).
If it wasn't clear, the question I'm trying to address is not:
It is this question (for which I think maybe):
And in addition (I try to give a tentative guess to establish plausibility):
You're proposing a third, important question (this I don't know the answer to):
- Are there investment strategies, besides index fund investing, that make the amount of money you have, your level of political access, and your accumulated knowledge infrastructure, relatively unhelpful, or even actively harmful, for evaluating certain specific types of investments - even if those resources are also extremely helpful for accessing and evaluating other trades?
I don't understand this sentence?
- In other words, are there investment strategies where individual, 'one-off episodes' of rational thought - as opposed to accumulated rational reflection over long periods of time - is more important than any other factor for accurately predicting price change for a subset of possible trades?
... still lost
- In other other words, are there investment strategies that are unattractive to hedge funds, but work well for small, smart, creative, and hard-working analyst-investors?
Fine, interesting question, I don't really see how it links to what you wrote in your post? You seem to posit reasons why strategies might exist, but really this is a concrete problem. "Name some strategies".
And in addition (I try to give a tentative guess to establish plausibility):
- If these investment strategies exist, can we identify them?
- How easy is it to end up with a compelling, yet wrong answer?
You're proposing a third, important question (this I don't know the answer to):
- If we can identify such a strategy, how many such opportunities can we expect to find, and how much competition will we face from other investors in our reference class?
I don't really see where you ask how to identify strategies? Nor whether or not you could delude yourself?
I don't really think LW is short of people suggesting strategies for small investors:
You seem to posit reasons why strategies might exist, but really this is a concrete problem. "Name some strategies".
So I think our disagreement might be this:
The reason that I think articulating why strategies might exist is that I'm a dogmatic EMH fundamentalist. When a person gives investment advice, I have historically ignored it, the same why I ignore it when somebody makes an argument for why I should consider using heroin, or committing suicide. Those behaviors are on my "no" list. I don't intellectually engage with arguments in favor of them, and short-circuit the updating of my priors.
Likewise, the EMH has put investment advice (aside from "invest in index funds") on my "no" list. I follow an iron law of not engaging with investment advice.
In concrete terms, this "iron law" plays out a bit like this:
The epistemic helplessness/fundamental trust problem here then precludes me from even starting to listen to Bob in the first place.
Hence, in order to ever get around to conceiving of, or even considering, a specific investment strategy, it's first necessary to argue against this pattern of thinking. That's what I'm trying to do here.
Hopefully, this illuminates why I don't consider it useful to try and articulate specific investment strategies, without addressing this more fundamental question first. My prior on any public investment strategy beating the market is so low that it has precluded me from even going there.
To change that, I need to understand why my priors should be high enough to consider specific investment strategies in the first place.
The reason that I think articulating why strategies might exist is that I'm a dogmatic EMH fundamentalist. When a person gives investment advice, I have historically ignored it, the same why I ignore it when somebody makes an argument for why I should consider using heroin, or committing suicide. Those behaviors are on my "no" list. I don't intellectually engage with arguments in favor of them, and short-circuit the updating of my priors
Likewise, the EMH has put investment advice (aside from "invest in index funds") on my "no" list. I follow an iron law of not engaging with investment advice.
I would highly recommend reading the full introduction to Section 20 of Cochrane's "Asset Pricing". (The whole course is excellent) Roughly he takes you through the progress academic finance has made since the 1970s, whilst not repudiating EMH, finding reasons why "invest in index funds" isn't necessarily the whole story for investors.
Thanks, this sounds very interesting. I appreciate all your thoughts in this comment thread and the recommendation as well.
Can you describe to me a concrete trade which looks like: 20% return on $1000. (All the ones I can think of tend to be just as amenable to professionals). The other issue of playing in the "micro-investment" pool, is typically liquidity is much lower, so costs are higher.
Here is one example. About a month ago I bought the stock ASK:DSK. The daily trade volume is $AUD100-200k. That made it easy for me to buy $10k worth. It is now up 48%. My slippage was minimal - I was able to buy it all at the offer or better with no market impact.
For someone with $10B FUM keeping the portfolio size to 100 stocks would mean that assuming they only bought 10% of the daily trades per day it would take them 3-4 years to buy in and a similar time to get out of the trade. So this stock, which is by no means unusually small, having a market cap larger than 2/3 of the stocks on the ASX, is out of reach for the big guys.
https://au.finance.yahoo.com/chart/DSK.AX
You've given an example which is already 10x what I asked for, and you could have plausibly done another 5x your size... I'm glad you made some money, but I don't think this is what I'm talking about
Feels a lot to me like a similar argument to why there is a complex ecosystem of investment sources for startups that target different stages of company life cycle: angel investors are "small time" and can afford to take high variance bets on fewer companies they know well while large VC firms have lots of money and need to find enough and big enough places to put it all.
On the other hand, there's some suggestive evidence that seed-stage returns have a power-law distribution with - implying that the best strategy is to filter out the obvious duds and then invest in literally everything else.
Epistemic status: Not investment advice. Just pure speculation by a non-rich investor in index funds and EMH partisan, inexperienced in the details of financial markets, doing his best to fit a simple model to a complex problem and steelman the opposing view.
One explanation for why you should invest in index funds goes something like this:
"You're competing against hedge funds, physicists, and the most rococo of high-frequency trading algorithms. Good luck, sucker."
I want to give an explanation as to why this might not be a complete perspective. It's based on two advantages of a small investor relative to a large hedge fund or big-money investor:
Inertia vs. agility
Professional investors are often managing huge funds. Bridgewater has almost $100 billion. It employs about 1,500 people to manage that money, so each employee is responsible, on average, for about $66 million. That's about average for a large hedge company.
There are about 630,000 publicly-traded companies in the world. If Bridgewater was looking at all of them, each employee would need to keep track of about 420 companies, giving each one just under 1 day of attention per year.
Realistically, hedge funds and their associated businesses have to specialize. And indeed, they do. They invest in forms of capital to support their monetary investments. They do things like using satellite imagery to monitor Walmart parking lots.
This gives them a gigantic advantage over individual investors. However, it also means that they are less agile, just like any other large company. The small time investor, not having spent a lot of money to analyze a certain company or market, may also be more able to explore new and strange markets, such as cryptocurrency.
Here, they may be more on an even playing field with the smart people in the hedge fund, who may not yet have figured out how to make sense of crypto. It's more a problem for individual human minds, not an experienced and organized collective of professionals.
Investment size floors
The huge size of a hedge fund, and the sheer amount of money that each employee is responsible for (on average) means that they need to find big trades to make the kind of profits that will pay for their fees, and deliver the returns their investors expect. Fees aside, Bridgewater needs to make about $10 billion per year to keep up with the market as a whole in an average year. Each employee needs to make on average about $6.7 million per year, or almost $25,000 per weekday, $2,000 per hour if they're working 12 hours a day.
And that's to do only as well as an index fund.
These are restrictions on the kinds of investments they can consider. They need to find not just investments that will be profitable, but investments where enough shares are being bought and sold to absorb the kind of money they want to invest.
By contrast, small-time investors can afford to consider companies that hedge funds may not be able to bother with, due to the smaller trading volume. When investing in these sorts of companies, they are most likely not competing against hedge funds.
Imagine that the hedge fund employee and the small-time investor (playing with, say, $1,000), are looking for opportunities to make a 20% return. The hedge fund needs to find an opportunity that can absorb $6.7 million, or many such opportunities to absorb smaller amounts. By contrast, the small-time investor only needs to find an opportunity that can absorb $1,000. That lets them cast a wider net.
To be sure, a small investor still must compete against other smart, small investors for whatever edge there is to be found. But now we're basing the EMH, with respect to these small-time and strange investments, on the intelligence of people more or less like you and me.
You might think that any thoughts you have on Apple and Walmart are light-years behind those of the hedge funds. But do you think your thoughts are light-years behind those of the other investors in a company with a trading volume of 1,000 shares/day? It seems far less intimidating to conceive of out-thinking the competition in an arena like that.
Implications
It's still hard to determine when you'd want to take the plunge and invest, outside of doing it as a sheer hobby. After all, earning 20% returns per year on $1,000 makes you a grand total of $200 before taxes, so I hope you didn't put too much time into finding that trade. And the more money you hope to invest in a given trade, the less the "investment floor size" argument above applies.
So perhaps this helps to partly contradict this argument.
On the other hand, perhaps you can, with experience, reasonably find investments that can absorb $10,000 with an average return of 20% with about 20 hours of work. That would yield a nice $100/hr for your trouble. That hypothetical Bridgewater employee would need to find one such trade per hour to make it worthwhile to consider such trades, meaning that they'd need to be 20x as efficient as you in identifying them.
Perhaps our markets are adequate for large trading volume stocks, but inadequate for low trading volumes.
These are tentative conclusions, with no deep understanding of finance standing behind them. Yet if it's worth doing with made up statistics, it's also worth doing with made up causal models. I look forward to your feedback, mainly in hopes of people who are better informed than me.