Years ago, I wrote an article titled “The Derivatives Myth.” In that article, I sought to address an ongoing market myth regarding there being this huge mountain of derivatives which would one day surely bring down all the banks (or some such apocalypse).
I did this by explaining how the very large (trillions upon trillions) often-quoted derivatives exposures came about. Basically, every trade that a bank does with a different counterparty adds up to more gross derivatives exposure. This is so even if the trade with one counterparty exactly cancels the underlying asset exposure taken through a trade with another counterparty. This obviously makes for huge reported notional derivatives amounts outstanding, even though the net exposure to any given asset can actually be quite small.
In spite of me and others explaining this, the same issue keeps coming up. Take, for instance, this Doug Kass (for who I have the strongest respect) piece yesterday. It again cites the derivatives book:
If all this wasn't enough, Deutsche Bank (DB) has an opaque derivatives book - probably at about $40 trillion with an estimated net exposure of approximately $100 billion.
$40 trillion sounds a lot, right? Even $100 billion sounds a lot, right? But now consider how the $40 trillion comes about. I’m going to make a theoretical example here:
- Say the bank sells a customer a derivative giving him $2 billion in long EUR/USD exposure (thus, the bank gets $2 billion in short EUR/USD exposure). Then, the bank sells another customer a derivative giving him $1 billion in short EUR/USD exposure (thus, the bank gets $1 billion in long EUR/USD exposure).
- As a result of the above trades, the bank would report $3 billion in gross derivatives exposure from these two deals. It also would report $1 billion in (short EUR/USD) net derivatives exposure.
It already can be seen that just two deals immediately start inflating gross exposure, and net exposure is lower. But it doesn’t stop there, beyond having many more customers putting on EUR/USD positions, you’ll also have others putting on exposure in other correlated assets. Let’s continue the example with the following:
- Now, the bank sells a customer a derivative giving him $2 billion in short GBP/USD exposure (thus, the bank gets $2 billion in long GBP/USD exposure). Then, the bank sells another customer a derivative giving him $1 billion in long GBP/USD exposure (thus the bank gets $1 billion in short GBP/USD exposure).
- As a result of this, the bank gets another $3 billion in gross exposure and another $1 billion in (long GBP/USD) net exposure. Gross derivatives exposure is now $6 billion (this adds up fast) and net exposure is now $2 billion.
But now look closer. Although the bank has net exposure of $2 billion, $1 billion is short EUR/USD and another $1 billion is long GBP/USD. EUR/USD and GBP/USD will show significant correlation between themselves. For instance, the EUR/USD-GBP/USD correlation can be 0.70-0.90 over time.
What this means in practical terms is that if the short EUR/USD exposure produces losses for the bank, the long GBP/USD will likely offset most, all, or in excess of all those losses.
Said another way, even the net derivatives exposure greatly overstates the actual risk the bank is taking. In a way, only the bank is in a position to evaluate the risk it’s truly bearing from the aggregate of all exposure it has. This is a very complex and largely automated task. It ultimately sees the broad day of light under the concept of “VaR” (Value at Risk).
VaR is based on all the exposures the portfolio has, the correlations between those exposures, as well as the volatility on each exposure. It's an estimate of the maximum the bank might lose on any given day to a given confidence level (usually 95%).
More On Deutsche Bank
The above was the main theme of this article. However, Doug Kass also seems to have been infected with the “EU banking is doomed” virus. Hence, there are other areas where Doug Kass didn’t properly look at Deutsche Bank. For instance:
- Deutsche Bank is reporting profits, not losses.
- Deutsche Bank’s leverage isn’t large considering the assets the leverage is made up of. Indeed, these assets being lower risk is what allows for higher leverage, as they contribute less in terms of RWA (Risk-Weighted Assets) to be counted against the equity base.
- Of course, low-risk assets got their yields pounded into the ground both by ECB’s ZIRP (Zero Interest Rate Policy) and QE (Quantitative Easing, buying those very same low-risk assets over and over and even at times pushing their yields into negative territory).
- The above is important. The low profitability and the leverage are connected. And as it turns out, both would be fixed over time under a higher interest rate environment. Namely, an end to ECB’s QE (this December) and ZIRP (6-9 months or so down the road) policies. The very policies which arguably made Deutsche Bank so unprofitable, given the way it had to concentrate on those “punished” (yield wise) assets, now stand to end.
- Also, Deutsche Bank's revenues, which Doug Kass says are low, are affected by the same dynamics. The huge mountain of low risk assets yields little. This (low) yield is part of revenues and part of NIM (net interest margin), which is the main contributor to profitability on the banking business, and which is very hard to attain when deposits cost zero but what the deposits are invested in yields zero as well.
- On the toxic loan book, this isn’t evident at all right now. There aren’t any obvious countries near default right now in Europe. The loans, having been toxic, affected Deutsche Bank and many other EU banks in the past, but those have been in the process of being cleaned up over time in many places, from Portugal to Italy, never mind the more central countries. If anything, Deutsche is probably too shy to put some of its assets to work in Italy, where yields are interesting right now due to all the fears. Italian banks will have it easier there.
- All in all, I see Doug Kass’s opinion as being informed by the currently negative sentiment on EU banking, and not so much on the developments which have been taking place in the last few years in EU banking. Sure, for something like Deutsche Bank, it’s harder, since low-risk assets are (still) yielding very little. And Deutsche Bank can’t shift heavily from those assets to direct loans on account of its equity base – so, Deutsche Bank needs to wait for those assets to yield more, which it will.
A final word on derivatives. The notional derivatives amount is ascertained because it sums up exposures to all different counterparties. With a $40 trillion book, there’s no way on earth that Bank of America (BAC), Citigroup (C), and JPMorgan (JPM) don’t have a large counterparty exposure to Deutsche Bank (which was a point Doug Kass also made, but to the contrary).
However, that exposure is worrying neither for Deutsche Bank nor for the U.S. banks. Those derivatives will have associated margin accounts which are updated daily and thus can’t easily produce a massive loss out of the blue (though large individual losses not entirely covered by margins can occur, with other minor counterparties).
I find Doug Kass's take on Deutsche Bank to be somewhat out of step with the current reality and future prospects. This is so even though Deutsche Bank remains near-term challenged, and those challenges only got deeper with much more unfavorable markets impacting some of its segments.
I especially find the take on the derivatives as continuing to further a myth. I hope my explanation of how those numbers come about, and thus why they are a lot less scary than they seem, helped some of my readers here.
A final note: Were Deutsche Bank to achieve its objectives for 2021, namely a 10% return on tangible equity, and DB's upside would quite likely be around 200%.
Disclosure: I am/we are long DB.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.