I have the feeling that we have an epidemic of stress, unhappiness and anxiety in the tech industry. It may be due to the working culture, perfectionism or simply because of toxic working environment. You could tell that from the suicide of Facebook employee a couple of weeks back, and Amazon employee in the last year. From time to time I would also experience such kind of tide in emotion in life. In order to tackle this I have developed my own kind of system to keep my mental health in control. And I didn’t find a good way to describe it by metaphor to explain it to another engineer until I am more familiar with the terminology of Site Reliability Engineering.

First, we have to identify catastrophic patterns to cause your system downtime. In life, it means you go depressed, and want to kill yourself, or simply lack of motivation to anything. One important observation is that the stuffs you might consider important are correlated to each other. The life could give you a straight of bloody Marry due to this. One bad thing could lead to the other. Even you are born in a middle-class family and have a good enough income, or slightly better in recent year as a software engineer. Your life could be easily destroyed. It doesn’t have to be financially, if you are depressed and committed suicide even you have billions in bank it is still a checkmate to your life. For example, it could be your marriage affects your work performance and get you fired, and you lose your income. Or it could also be your important partner is sick, it not only costs you a fortune to pay the medical bill if you are living in countries like U.S.A or Singapore, where medical bills are expensive comparing to the median incomes. And since you have financial burden, your significant partners could be tempted to abandon you since they would not like to be involved etc. In order to prevent the chain reaction, you have to manage your “components” to be more resilient against each other, and not be over-confident to have too many components in your life. For example: getting overly leverage financially on your housing but at the same time raising kids, after all raising kids and marriage are liabilities that is off balance sheet. And if you already have components you are reluctant to get rid off, make sure you have insurance covered in a proper way to reduce the over all risks, at least stopping the chain effect when it happens. The best, for sure, to have low liability.

Once you find a way to manage your portfolio of liability, a portfolio of “bomb shelters” or “chambers” where you could find peaceful time to recover is important. Over-stretching on a specific goal is dangerous since if that goal fails your mental health would be seriously damaged. For example, if you would like to climb the ladder of corporate or build a company. You have to go all out but make sure not to over stretch that could cause a total meltdown. If your goal fails, if you still have other goals like playing sports/games where you could find small achievement, that’s a hedge for your mental meltdown when you fail to build a company or climb the ladder. It’s a diversification of your goal, to make sure it’s unlikely that they correlate each other. If your secondary goal is to tied to your kids or partner, due to human minds are not stable. They might not care or understand your situation when you are down for those goals. It could be a shelter but not a reliable hedge. Having nice food is a good one since it’s unlikely that it would be affected by human factors, but having it too often it could hurt your lookings or physical health as well.

Other than the diversification, I find stoicism is also useful to me. Here is a video from Tim Ferriss.

I am not a fan of what’s been in American main stream culture on the alpha-male like of “Leadership.” Human are not infalable and actually quite easy to be broken. Overly stretching to the image of “confidence” and “overcome your poster syndrome by pretend it until you make it.” would easily make yourself vulnerable to a strike-back. I believe the Newton’s third law of motion applies here as well, if you stretch too much, you are exerting the same magnitude of force in the opposite in the same time.

These are the “tips” or “speculation” that I have in mind for years. It’s not a science or corollary that would be deemed as universal and could be applied any kind of extreme situation (for example, war time or you are put in concentration camp). But at least it kind of works for me in modern time living in big cities.

前一陣子因為 The Big Short 而出名的 Michael Burry 回覆了 Bloomberg 對於市場看法的信而又引發了網路上一場論戰。因為他的回覆中關於 Index Fund 的說法提到了 “Bubble” 這個詞而許多人擺錯了重點，把重點擺在是不是有 Bubble 這件事。其實他的說法跟之前 Howard Marks 還有 Seth Klarman 提到的並沒有不同，其實關鍵的重點在於結構設計，而每個結構設計跟實體工程一樣都有他的極限在，特別是在金融市場中因為結構承重量或容錯量是不可見的，涉及抽象的概念。雖然我完全同意 Micael Burry 的說法，但我一直想不到一個好的比喻跟一般人解釋這件事。直到我前幾天看到 yinwang 關於自動駕駛責任的文章。我覺得解釋得非常清楚，而同樣的邏輯謬誤也是可以套用在市值權重指數基金擁護者的回覆上。

要討論這件事涉及兩個層面

1. 為什麼用【市值權重指數基金拿到的是市場報酬】不足以回應 Michael Burry 點出的風險？
2. 市值權重指數基金的缺陷在哪？什麼情景下會觸發？

首先用一個例子來說明第一點。在今年初發生了 Boeing 737 MAX 因為設計不良而造成的事故。一時間人心惶惶，而各國政府也相繼搬出政策禁飛 Boeing 737 MAX。為什麼大家會擔心這件事？不是說 “平均” 來講，發生空難的事件極低嗎？ 機率上來講你死在去機場的路上是遠高於你發生空難的機率。這主要的問題是因為 Boeing 737 MAX 有共同的因素驅動，也就是設計的缺陷。

就機率來解釋的話，就是

P(其它 Boeing 737 MAX 墜機  | 一架 Boeing 737 MAX 墜機)

高於

P(其它非 Boeing 737 MAX 墜機 | 一架非 Boeing 737 MAX 墜機)

因為他們不是類似於獨立事件。而是有共同的缺陷。假如每一家航空公司全部都是提供 Boeing 737 MAX 的話，說我身為一個個體明天搭的飛機墜機的機率趨近於平均發生空難的機率一點意義都沒有，因為平均的結果是大家一起因為起飛後因為設計的缺陷造成機頭直接往下墜毀死亡。

所以說【市值權重指數基金拿到的是市場報酬】並不足以回應。正確的回應要解釋為什麼市場報酬背後沒有共同驅動因素造成大家一起死的結果。這在大多數時候是對的，畢竟市場報酬是買方跟賣方資金平衡投票出來的，而且市場中每個人的想法不同，就好像是不同設計的飛機一樣。所以大多數時候市場的驅動的因子沒有共同背後的因素驅動。但接下來解釋的第二點，我會說明在某種特殊情況底下是有可能發生的，也就是市場上的參與玩家是同一個想法佔大多數的時候，而指數權重基金佔大多數也就是其中一個實現。

要了解第二點必須說明市值權重指數基金的運作，這邊舉一個簡單的小例子說明。台灣人很愛買房子出租，所以用房子來舉例，

今天市場中有三種分十年折舊的房子可以帶來租金收入，然後市場裡面有四個投資人 A, B, C, D

期間是 10 年，十年後房子價值歸零。市場無風險利率是不會動的 1%，而對於房地產的投資人 A, B 風險利率要求是 2%，加上通膨 3% 折現率共 5%。那這些房子在他們兩個人間的平衡價錢應該是多少呢？用十年的 DCF 可以得到

而投資人 C, D 決定採取不一樣的觀點，不用類似 DCF 的方式衡量，他們決定編制 FOOBAR 市值權重指數，這個指數就是

i 的市價 * (7.72 / 100.38) +
ii 的市價 * (15.44 / 100.38) +
iii 的市價 * (77.22 / 100.38)

他們把他們的錢加起來成立市值權重指數基金，把他們共有的錢 100.38 照上面比例分配買。也就是 i 買 7.72 塊， ii 買 15.44 塊， iii 買 77.22 塊。所以投資人 C, D 認定的價值不是根據衡量物件本身，而是因為市場中其他人的價錢而分配他們的錢的分配。假如今天 A, B 覺得 iii 其實收不到 10 塊的年租金，其實只能收到 2 塊。那就會變成 15.44 塊。那 C, D 就會跟著把 iii 賣掉。因此你可以看到一般所謂的市值權重指數基金就是一種 Trend Following 的策略，是因為市場中其他對價值做出反應的人的定價而跟隨。在這個例子中，當 C, D 反應過來的時候，他們賣的價錢可能會低於 15.44 塊。因為他們的對手只有 A 跟 B。而 A, B 只願意用低於 15.44 塊的價格收，不賣的話就違背了他們指數的規則，所以他們一定要賣，這是機械性訂死的。 為了要賣掉他們就不得不降價，這邊的例子只有 C, D 兩個個體的資金，但你可以想像賣方是這個的 100 倍, 1000 倍然後【同時】想要賣的情況，也就是會造成 race to the bottom。最後平衡價可能遠遠低於 15.44，也許 7, 8 都有可能。

但市場有趣的地方就是上述情況又不是鐵定會發生。假如今天反而是 A, B 突然都變調，也投入 C, D 的指數基金。市場中就不再有人用 DCF 去定價了。所有人都認為權重永遠都是如下：

i: (7.72 / 100.38) 
ii: (15.44 / 100.38)
iii: (77.22 / 100.38)

整個 price discovery 的機制就不見了，不管底下的實際能收到的租金有多少，就是這樣定價，所以說股價都是所謂人類基金管理者亂炒作這說法並不公平，市值權重指數是一種跟隨策略，沒有人類基金管理者他也是不會動的。但這並不是一個穩定的平衡，遲早有人會跳出來點出國王的新衣，然後前面崩毀的情況就會發生。

從上面的邏輯推導與立論可以知道，市場大多數人都在使用某個系統時，而那個系統又因為一些背後共同問題造成時，平均就是大家一起平均死。沒錯，你拿到平均的結果，但你還是死了。但同時，請不要滑坡推論就說 Boeing 或是市值權重指數基金就是該死，身為渺小個體我們應該完全摒棄，因為

1. 統計上 Boeing 跟 Airbus 造的飛機比起你自己亂造的飛機要強，儘管 Boeing 737 MAX 有問題。
2. 統計上 Boeing 跟 Airbus 造的飛機比起你自己請的飛機廠商也要強，儘管 Boeing 737 MAX 有問題。

但反面來說

1. 統計上有瑕疵的立論並不能因此讓我不擔心 Boeing 737 MAX 的結構性問題，除非你能舉出有說服力的邏輯推論說明我舉出的情況不用擔心，因為儘管機率不高或是未知，但我一旦中了就死了。同樣對於市值權重指數基金也是，這關係到許多人畢生的積蓄。

我想聲明，市值權重指數基金是非常偉大的發明，他讓大多數不懂金融的人都有機會分享到果實，能承受的資金量也非常巨大，至今都沒發生問題。

我並不是鼓勵說就不要使用指數基金，但與此同時，因為過去沒發生問題而認為以後也不會發生問題，不去用正確的邏輯回應問題。這種過度膨脹的心理而不去理解市值權重指數基金的結構所潛藏的缺陷並提醒大家，在我看來跟那些設計各種金融工程工具但不去點出工程極限的人沒有兩樣，都是不負責任的。

最後安插一個軼聞。Michael Burry 在 2000 年就成立 Scion Capital 來管理家族基金，但他也接受外部他喜歡的人的投資，早年完全是採取類似 Ben Graham 或是 Warren Buffett 在 Partnership 時代的策略，很多人是因為 The Big Short 才認識他，但不知道他是跟 Warren Buffett 同一風格的投資人。好巧不巧在 2001 年的時候，市值權重指數基金的教父曾經在 Forbes 專訪上嘴 Michale Burry。

In the 2001 article, Forbes proposed the following as 
a working definition of a hedge fund: "A hedge fund is 
any investment company that is unregulated, has limited
redemption privileges, and charges outrageous fees." 
The author also enlisted the help of John Bogle, the 
Vanguard index fund czar, to make the case against hedge 
funds. Bogle obliged, noting that it is not realistic 
for investors to expect the hedge fund industry of more 
than 6,000 funds and $500 billion in assets to outperform 
the rest of the market over the long term.

This is a reasonable observation, and I don't necessarily 
disagree with him. Unfortunately, he then went on to pick
a name at random from a hedge fund directory to disparage, 
saying: "I don't know what to do about Scion Capital, 
started by Michael Burry M.D. after leaving his third year 
of residency in neurology.  He started it mostly with his 
own money, $1.4 million, and he's looking for more. His 
technique to manage risk is to buy on the cheap and, if 
he takes a short position -- I hope you're all sitting 
down for this -- it is because he believes the stock will 
decline."

很可惜，他完全就挑錯了人來嘴，我們都知道 2008 發生了什麼事。