Martin Shkreli, Surprisingly Good At Fraud

From Bloomberg's Matt Levine, first he sets the table:

There is a fairly standard trajectory for a lot of securities frauds. A brash youngster comes along and thinks he will be good at trading. He raises money from people who also think he will be good at trading. Maybe he lies to them a bit -- about how good he already is at trading, or about how much money he's already raised -- or maybe, at this stage, everything is completely on the up and up. In any case, he gets the money and starts trading. Then he starts losing. This is bad. It's bad for his bottom line, of course -- he's getting paid to invest his investors' money well, not badly -- but more importantly, it is bad for his self-conception. In his own mind, he is a winner; he is supposed to be good at trading. It is unbearable to him that he isn't.

So he starts lying, telling investors that everything is great and he's making money for them hand over fist. This is a hard lie to sustain, since eventually the investors will want their money back, and he's promising them more and more of it. There are only two real ways to sustain the lie. One is to Ponzi it up, and keep raising new money from investors to pay out the old ones. In the right environment, and with the right sales pitch, this can work really well for a really long time, but since the supply of suckers is finite, it can only ever end in prison or the grave.

The other way to sustain the lie -- and the only possible permanent solution -- is to gamble on redemption. If you've lost half of your investors' money, but you've told them that you've doubled it, then you need to quadruple the money you have left. So you take riskier and riskier bets, hoping that one will eventually pay off and make up for all of the earlier losses. This basically never works! But it could work. And if it does, your investors will be happy, you will be rich, and no one will ever be the wiser....Then MUCH MORE