Item — HN Reader

Mathematically speaking though, Michael Berry's success may have been more to do with luck than skill:

1. Nowadays, options are priced in such a manner that the expected value of buying OTM hedges is almost always negative. This is due to the inflated volatility for out-of-money put options and other factors.

2. It just so happened that the subprime mortgage meltdown coincided with his career; for the period between 1935-2007, his strategy would not have worked. It also would not have worked between 2008-present. (maybe it worked have worked in 1987, so we're talking three instances out of a century). Right now, the evidence suggests bank have reformed their lending practices, with higher credit scores for new homeowners, so the odds of another financial meltdown are slim.

3. The fed balance sheet is big, but interest rates are very low.

The fed has posted a large profit , a 30% gain in 2015 for a profit over over $100 billion, which is sent back to the treasury

http://money.cnn.com/2015/03/20/investing/fed-profit-balance...

And the evidence due to the mathematics of the yield curve and the composition of the fed's holdings suggests the the odds of the fed losing money on it holdings is very low:

http://blogs.wsj.com/economics/2011/04/11/odds-of-fed-losing...

“Short-term interest rates would have to rise rapidly to quite high levels — in the neighborhood of 7% — for the Fed’s interest expenses to surpass its interest income. Such an outcome appears very unlikely,” the paper said. In the event that the Fed did face a loss, it could simply hand no money back to the Treasury and, “in the most extreme case, future remittances would also be reduced (and recorded as a change in deferred credit), but the Fed’s capital base and financial position still would remain completely secure.”

Japan has a much bigger debt, they seem to be doing fine. Low interest and srong dollar is due to reserve currency status, flight to safety, emerging market weakness, commodity weakness, petro dollar, the large size of the US economy, and other factors.

3 comments

I was lucky---when I was born in 1930, I won the ovarian lottery. You know, I was born in the United States. The odds were 30 or 40 to one against that. I was born white. If I had been born black, my life would have been different. I was born male. My life would have been vastly different if I had been born female.

So, I had all kinds of lucky things.

-Warren Buffett, 4 Oct. 2011 [1]

Certainly Michael Burry had enormous luck. But there were plenty of others with the same amount of luck who did not see the crisis coming or who did nothing about it.

[1] http://fortune.com/2011/10/04/warren-buffett-at-fortunes-mos...

With regards to 1: So?

Buying options is buying insurance. From the point of view of the purchaser, the expected value of buying insurance is always negative (otherwise insurance sellers would be out of business). However, insurance is not bought due to its EV, it is bought due to its variance-minimizing properties.

Not really, insurance is bought by people because of the non-linear utility function of money: losing $1 million has a far more negative utility that 1000 times the utility of losing $1000.

The same with lottery tickets. People don't buy them because they're stupid, but because they value $100M far more than 50M times the value of $2.

I love that you think there is one and only one reason to buy insurance. I suspect that people buy insurance for both reasons. Which is to say that you're both right, and both a little bit wrong.

Oh no, I don't think there's ever a single reason for anything. People don't buy insurance to reduce variability, since people don't care about positive variability, only negative (as in the lottery example).

Some people buy just because other people tell them to, some because they are afraid of losing something, and so on, there are a thousand reasons why individuals do it. But the general trigger of insurance for people is the non-linear utility function.

> But the general trigger of insurance for people is the non-linear utility function.

I would respectfully disagree. Homeowners insurance is a great example. Most people buy it and never make a claim, and they "lose" insofar as their premiums are gone for no monetary return. But some people who buy it "win" and get to collect enough money to roughly replace their house, belongings, etc when a hurricane, tornado or whatever destroys it all.

Not all insurance is the kind where you pay the same amount whether you make 1000 small payments or one big one.

"Not all insurance is the kind where you pay the same amount whether you make 1000 small payments or one big one."

But I never said that! It's just like the lottery, the payout of the lottery is smaller than the price divided by the odds (that's why a lot of people call it a tax on people who don't know math), but as the utility function of money isn't linear, for most people a small chance of gaining a lot of money is more valuable than a little bit of money.

The same is valid for insurance. The value of losing a lot of money is far more negative than the cost of it, even when taking the risk in account.

That's because the utility function doesn't depend only on the gain, but also on how much you have. A billionaire might not care about home insurance for a 100k shack.

Even if the utility function was linear, I'd still prefer to insure some things, even with a negative expected value, for the sake of smoothing variance.

EDIT: I'll clarify this to mean; even if the EV of insurance was exactly neutral, I'd still buy it for smoothing. And even if the EV was negative, I'd still buy it for smoothing. There's a difference between smoothing and loss aversion.

> The fed has posted a large profit , a 30% gain in 2015 for a profit over over $100 billion, which is sent back to the treasury.

Yes, those are profits on assets they have on their balance sheet. How much QE had to be done to achieve that profit. It's like the invisible negative side of the balance because the cost is socialized and delayed by years, by the time the damage is done few will correctly identify it as the dominant source of the problem.

It's easy to post a profit on select assets when you're pumping it up with the other hand.