In general, Wall Street hates negative carry trades. But it’s likely that nowhere were negative carry trades more loathed than at JPMorgan Chase. Ina Drew, the firm’s former chief investment officer, who left the firm on Monday amid the trading scandal, reportedly believed that the bank could hedge against business losses and still make money at the same time. That’s very hard to do in general. But it’s impossible to do with a negative carry trade. That’s because, until they payout, which not all do, negative carry trades cost more and more money the longer you hold them.

Most negative carry trades involve buying insurance and paying a regular premium. But they don’t have to. If you rent, because you believe housing prices are going to drop and that you will be able to buy a home cheaper later, in Wall Street speak that’s a negative carry trade. The rent you shell out each month, minus what you would have paid in interest (after-taxes) on your mortgage and property taxes, is your negative carry. And the longer you rent, the more housing prices have to drop to make your choice to wait pay off.

That’s not to say negative carry trades are always bad. Some have been spectacularly profitable. John Paulson’s bet against the housing market in 2006 and 2007, which reportedly netted $25 billion, was a negative carry trade. But if you are running a trading operation at a big bank, and you are watching your P&L everyday, like Drew reportedly did, to make sure your hedges are as profitable as they can be, you will do whatever possible to avoid negative carry trades, even if doing so opens you up to massive losses down the road, which it appears is exactly what happened to Drew and JPMorgan.

MORE: Tossing blame for JP Morgan trade? Don’t forget the Fed.

The bet that blew up in JPMorgan’s face probably started mid-last year, but it could be even older than that. Banks, especially large banks, are generally betting on the economy all the time. They give out money to people and businesses in the hope that they will get paid back with interest. The problem is that in times of economic stress, the business of banking is not always a good bet. But you can’t close the doors. So, if you are big bank, what you do is hedge.

The easiest way to hedge your bets these days is to buy so called credit default swaps, which are essentially insurance contracts that pay out if a loan goes sour. That’s exactly what JPMorgan started doing in mid-to-late 2011 as the economy started to slow, Washington gridlocked and the problems in Europe grew. JPMorgan appears to have bought insurance against a number of large U.S. corporations, protecting the bank against the possibility that if the economy did fall into a double dip, as more and more people were predicting, the bank would be covered against the chance that some of its largest corporate customers would default for the next 18 months. The contracts were short-term and expired in December. And even JPMorgan’s borrower didn’t default, just the rising threat of higher defaults would likely cause short-term corporate bond prices to fall, and yields to rise, and make the insurance contracts JPMorgan was purchasing increase in value.

But while that hedged the bank, the trade, like all negative carries, was also costly. Drew’s chief investment office lost $100 million in the second half of 2011, ending the year up just $800 million, compared to a profit of $1.3 billion the year before, and a gain of $3 billion in 2009. What’s more, the economy didn’t fall off a cliff, instead it started to improve and by February again looked well on the path to recovery. At this point, what JPMorgan should have done was close out its insurance bets, and take the loss. Or at least left them on and just swallowed the CDS premiums as a cost of doing business. Afterall, even with the costly trades JPMorgan was still able to turn in an overall profit of nearly $19 billion last year. The bank could afford to have some insurance.

MORE: Corporate America’s double standard on the deficit

But that’s not what JPMorgan did. Instead, Drew’s traders sold CDS contracts on an index that tracks the credit worthiness of over 100 U.S. companies called the CDX IG 9, for short. And they quickly sold a massive amount of insurance – reportedly as much as $100 billion in a few months – which is why the main trader who was in charge of putting on this part of the trade, Bruno Iksil, came to be known as the London whale.

To the outside world, it looked like JPMorgan was making a huge bet that U.S. corporate credit prices would rise, and long-term yields would fall, and the U.S. economy in general would improve. Given that is what a bank does normally, this didn’t look like a hedge. And so a number of observers, when news of the trade originally emerged a little over a month ago, raised questions as to whether JPMorgan was violating the Volcker rule, even though it really isn’t yet in place.

But JPMorgan’s CEO Jamie Dimon has repeatedly argued that the trade would be allowed under the Volcker rule because it was a portfolio hedge, which in a way it was. Unlike the CDS contracts the bank had bought just a few month earlier and held on to, the new contracts that it sold didn’t expire until 2017. So when you combine the competing short-term and long-term trades, JPMorgan now had a bet that the yield curve on U.S. corporate bonds would flatten, which is exactly what it would normally do if we were headed into a recession. And yet, because they had now sold a massive number of CDS contracts instead of buying them, JPMorgan was being paid on a regular basis to keep the trade going. The bank had turned its negative carry trade into a positive one. Not bad.

MORE: Not working: Voices of America’s laid off workers

Except, of course, instead of just taking out insurance on some of its borrowers, JPMorgan now had a massive bet on the shape of the yield curve. If it was to steepen instead of flatten, JPMorgan would lose billions of dollars, which is, as we now know, exactly what happened. Why that happened isn’t entirely clear. In part, the economy hasn’t double dipped. But what also happened is that a number of hedge funds spotted JPMorgan’s trade and began betting against the bank’s positions, putting pressure on its trades.

In the end, the real problem was the original fallacy that Drew set up, which is the idea that banks can both hedge their positions and make money at the same time. The financial crisis should have proven that this wasn’t possible. Unfortunately, it appears that Wall Street needed another reminder.