site banner

Do you believe in efficient-market hypothesis (EMH)?

During discussion about things where you had strong opinions and then changed your mind, someone mentioned EMH. Do you believe in EMH and if so is it strong or weak version?

I used to believe EMH but not strongly. The pandemic changed my view because I managed to invest some money when the stock market dived and was clearly influenced by overly pessimistic view of the impact of covid. Some might argue that the market reacted to the irrational government measures, so it is not that the case that the market was mistaken. I still think that investors were equally irrationally pessimistic. I reject the view that this is a hindsight and I was merely lucky. I am not big expert and I did not possess any proprietary information. I had the same information as everybody else, I just didn't let my emotions take over me. This is further confirmed that even today when all the events have passed exactly as predicted, majority of people still maintain their mistaken views that covid was very dangerous to young and non-risk population.

It is the only time when I saw the rest of the society to be so wrong in their views and clearly this was my once-a-lifetime chance. I haven't see any other opportunities for easy money so far but I think that people who are experts in their fields and investors might have been able to find more opportunities.

One of them was found by Michael Burry who definitely saw that the 2008 financial crisis was coming. He wasn't just lucky because he had read and analysed all the documents and had to create special investment instruments to profit for it. In this way, it wasn't easily accessible by laypersons like me who have no time or understanding about investment. Again, most professionals were blinded by collective frenzy.

What is your opinion about EMH?

3
Jump in the discussion.

No email address required.

You reasoning is largely faulty:

  • "I beat the market once, so the market is inefficient" is pretty obviously faulty. You might as well say "I beat the world champion poker player in three hands, so I'm obviously better."

  • Pointing to one investor who thought he saw the 2008 crisis coming is also not a knock-down argument. Even if you ignore luck in the sense of randomly getting an investment decision correct, it is still entirely possible that specific investors can see something coming while the market as a whole is efficient in the sense that it's not generally worth trying to beat it (even for the investor who beat it in this particular situation). [ As a toy example, suppose there are 1,000 factors to think about. Suppose if you thought of the correct 3, you could have seen the 2008 crisis coming. If you chose 10 random factors to think about. Your odds of seeing the crisis coming would be 1,384,725-to-1 and it is therefore not worth you spending the time to think about it.]

Broadly speaking, the EMH is largely correct within conventional liquid asset classes over short time horizons during stable economic conditions. It is frequently incorrect between illiquid asset classes over long time horizons in unstable economic conditions. There is a continuum between these two investment scenarios.

Concretely, if you're day-trading stocks during a boom, you're probably playing a losing game. If you're deciding between buying long-term stocks, bonds, or housing during a financial crisis, it is quite plausible that you can find significant relative alpha.

Pragmatically, the most valuable investment questions are typically related to avoiding taxes and choosing appropriate degrees of leverage given your age, life plans, risk tolerance, and interest rates. You can achieve several extra percentage points in risk-adjusted returns by carefully considering those questions.

When I learned that US embassy workers were having unexplained brain damage, many people ascribed that to “sonic weapons” from unknown attackers. The most likely explanation however was that it is probably coincidence and more likely it was mass psychogenic illness. This Havana syndrome is still unresolved but if there was a way to bet on this, the correct way would be to bet against sonic weapons.

Now, if the governments in many countries would wrongly become convinced that it was sonic weapon, it would seem that you lose the bet. But eventually the truth comes out and the bet against sonic weapons has much higher chance to be right.

Covid was something similar to this. When due to human nature collectively we have markets going into wrong direction, you may or may not be able to profit from it but in any case you need to take stance and try to correct the delusion that is harming all of us. If you can it directly by placing winning bets, that's preferrable because it might help other people to realize sooner that their thinking is faulty. Granted such cases are rare, or non-existent for most people.

To be completely honest, I'm not even really convinced you understand why you got your covid bet "correct".

The single most important factor in asset price movement wasn't the degree of government crackdown. It was the enormous and unprecedented degree of stimulus unleashed by the Federal Reserve and federal government. In particular

  • The Federal Reserve dropped nominal interest rates. Government stimulus drove inflation. Together, these drove real interest rates literally negative. Low interest rates encourages leveraging, which drives up asset prices.

  • The Federal Reserve implemented enormous quantitative easing, driving an artificial demand for assets, driving up their prices.

  • Government stimulus drove profits up ~25% in real terms, driving up stock prices.

Altogether, this explains why an initial 30% drop ended up becoming a 45% surge peaking near the end of 2021. Then, as soon as the Federal Reserve started hiking interest rates and reversing quantitative easing, we saw stocks start to fall.

Why did the market take so long to price this in? Because the actions of the Federal Reserve and federal government were unprecedented and really didn't make much sense. Why were rates kept near zero when inflation hit 7%? Why did stimulus checks go out when unemployment was below 4%? None of it made any sense and all of it was completely out of line with historical precedent.

The effect of biased predictions re covid lockdowns was dwarfed by the unanticipated actions of fiscal and monetary stimulus. If I'm right, then you really made the right bet for the completely wrong reasons.

Though, to be honest, even if I thought lack of covid lockdowns caused stock prices to increase and even if you predicted this ahead of time, I wouldn't consider that a serious mark against the EMH. Consider another toy example:

  • Investment return = sum of 5 dice: [3, 6, 2, 5, 4] = 20

  • You have access to 3 dice: [4, 5, 6]. You therefore extrapolate that the investment return will be 20 = (4+5+6)/3*5

  • The market has access to 4 dice: [4, 5, 6, 2]. They therefore extrapolate that the investment return will be 21.25 = (4+5+6+2)/4*5

You made a perfectly rational prediction. You did better than the market. And this is all perfectly consistent with the market being a better investor than you (notice they have a superset of the dice available to you).

This example is a toy. The numbers stand in for factors / arguments / evidence. You can get an answer right by virtue of having less evidence if you get lucky. From the inside, this feels rational. But, with repeated testing/betting, it is eventually revealed as luck.

Well, I had read Krugman's “End This Depression Now” explaining why stimulus in certain conditions is needed. I read the book because I was going through personally depressive period in my life but the book had some insights anyway. Krugman was proven right with sufficient evidence that I was surprised that so many experts still disagreed with him and demanded austerity policies instead. Nevertheless, even the staunchest critics have to admit that stimulus worked better during economic contraction, so it was no-brainer that the governments faced with economic contraction during covid pandemic would try to use stimulus too. I didn't believe that it will be very effective because unlike normal conditions people were prevented from spending money. The government had little choice though because they had to make sure that people who were forced out of work by their policies can still pay rent, buy food etc.

Incidentally, the lack of spending opportunities also gave me chance to invest some money. I expected that other people would do the same and it seems that eventually they did. But when I saw stock market taking a dive, it made no sense. I didn't expect to make a quick profit, I thought that recovery would take longer time. I believed that eventually everybody would get covid and then all the restrictions will be shown useless. Even the most totalitarian governments would have to admit that their policies are futile. Once restrictions are gone, people would quickly restart economic activities and stock market would rise.

Scott wrote that the stock market dive actually happened due to automatic selling. Many funds set automatic scripts to sell when the price reaches certain low threshold to prevent from further losses. Selling moved the price even lower with more and more funds initiating automatic selling.

My central point wasn't that you didn't know what stimulus was. It was that you ascribed the large stock returns to covid-related measures rather than unprecedented stimulus, and I believe this is wrong and demonstrates that you beating the market was, in fact, luck and not skill.

I'm not against government stimulus. In fact, I'd even consider it plausible that the central bank and federal bank have never used too much stimulus prior to 2020. But it should be obvious that they used too much in response to the covid pandemic.

Biden signed a $1.9 trillion stimulus package on March 11, 2021 - inflation was 1.7% and unemployment was 6.1%. For context, the inflation target is 2% and the median unemployment rate historically is 5.5%. In other words, the economy was only ever so slightly below average in business-cycle terms. Meanwhile, the stimulus package was larger than all stimulus spending in response to the Great Recession. And that's just the federal government. The Fed continued its quantitative easing and purchased over a trillion dollars in assets after March 11th, above and beyond it's previous purchase, all the while keep nominal interest rates at 0% (and therefore real interest rates negative). It is really hard to overstate how unprecedented the degree of stimulus was. This is why stock prices surged: the degree of stimulus was enormously unexpected and had enormous direct and indirect effects on financial markets.

I didn't believe that it will be very effective because unlike normal conditions people were prevented from spending money.

Stimulus can have three effects: employment, inflation, savings. I agree it was less effective at promoting employment and inflation, but this (again) goes back to my point: it mostly promoted savings - i.e. investment. But also it's not like stimulus was completely ineffective at boosting employment: unemployment dropped from 14.1% to 6.1% in 12 months, a feat literally never seen in US history.

The government had little choice though because they had to make sure that people who were forced out of work by their policies can still pay rent, buy food etc.

They had lots of choice. The most effective way to achieve their goals would have been to send checks to the unemployed. Increasing unemployment benefits is bog standard during a recession anyway. Sending checks to everyone and engaging in generic stimulus was not a forced moved on the government's part. But this is all besides the point. We're talking about financial markets, not whether stimulus is good.

I didn't expect to make a quick profit, I thought that recovery would take longer time. I believed that eventually everybody would get covid and then all the restrictions will be shown useless. Even the most totalitarian governments would have to admit that their policies are futile. Once restrictions are gone, people would quickly restart economic activities and stock market would rise.

Ok, but the question is NOT whether this line of reasoning is correct. The question is whether hedge funds et al had already priced it in. You have given no evidence indicating they hadn't.

Scott wrote that the stock market dive actually happened due to automatic selling. Many funds set automatic scripts to sell when the price reaches certain low threshold to prevent from further losses. Selling moved the price even lower with more and more funds initiating automatic selling.

Yes, this is due to forced unwinding of leveraged positions, which often results in short-term liquidity spirals. I'm not sure what your point is though?

You are missing the main reason why the stimulus was too much this time. It was simply because the government acted in contradictory ways and prevented people spending money. The fact that different restrictions continued after most elderly had received vaccine was the biggest unforced error.

It was simply because the government acted in contradictory ways and prevented people spending money

I'm not saying the restrictions were justified, but you are dramatically overstating their macroeconomic impact, especially post-vaccine rollout. In 2021-Jan, vaccines had just started rolling out and personal consumption had returned to its pre-covid peak. You think that's a bigger factor than the largest stimulus bill in history being passed during middling economic conditions? Not to mention monetary policy...

I am not sure I believe that. The reason of the crash and everything that followed was solely due to the government's mistaken actions. I don't know specifically about the US, but the life in 2021 was far from normal and it was also solely due to the government's mistaken actions. It would be strange if it had no deleterious effect on the economy.

The government was trying to whitewash the negative effects on people.

I never said it had "no deleterious effect on the economy".

I am not sure I believe that

The data is easily found.

More comments