Behind a contested headline is a complicated, somewhat philosophical question: we all know that JPMorgan has a very big trading business, but how big is it?
As reported last month by The Bureau of Investigative Journalism’s Josephine Moulds and David Kenner of the International Consortium of Investigative Journalists:
A JPMorgan whistleblower said the world’s biggest bank was misreporting its trading activity, which could have inflated its earnings by billions of dollars and added millions to executive pay packages.
They reported the issue in a letter sent to board members, the Federal Reserve and the financial regulator, the Bureau of Investigative Journalism (TBIJ) and the International Consortium of Investigative Journalists (ICIJ) can reveal.
Another banker familiar with the matter said other major US banks were doing the same — with tacit approval from the Federal Reserve, the US central bank.
“We are confident in our methodology, which is fully transparent to our regulators,” JPMorgan told FT Alphaville. “We comply with all capital regulations, and claims suggesting otherwise are false.”
The answer to the philosophical question is, of course, “it depends who’s asking”.
If you’re a manager who wants to know the risks your bank is taking, then you’re interested in the sum of the trading positions reported on a net basis, with short positions subtracted from longs. Depending on the sophistication of your modelling, you might even allow positions in different securities to partly offset each other if they are negatively correlated.
For other purposes, you might not be so generous. Measuring risks on a net basis is all very well in good times, but when the chips are down, net settlement often turns out to be trickier to achieve than one might hope. You might not want to allow positions to be netted if the long and short arm were held in different time zones, or if there were no legally enforceable netting agreements. You might at least want to be aware of the biggest possible gross number, with short positions added to longs.
And sometimes, you’re measuring trading positions not because you care about what risks a bank is running, but because you just want to know how big its trading book is. Because you care about how much everyone would miss it, if it were gone — effectively, how important it is to the global financial system.
That’s the sort of data the Basel Committee uses when it gives its rankings for the Global Systemically Important Banks every year.
And indeed, the relevant reporting standard says that:
Whistleblowers speaking to TBIJ/ICIJ allege that since at least 2018, JPMorgan hasn’t been doing this — it’s been netting longs against shorts “at the legal entity level”, so only those securities in which JPM had all their longs in a different subsidiary from their shorts have been reported correctly. (JPMorgan says it complies fully with all capital regulations and is confident in its methodology.)
Would siloing longs and shorts in the way described above make a difference? In theory, yes. A smaller trading business means a lower overall systemic importance score. The TBIJ/ICIJ report claims that by reporting a smaller book, JPM was grouped lower by one “bucket” in the GSIB league table than it would have been otherwise.
Every step downward is 50bp less of additional capital requirements; for a bank the size of JPM, this is equivalent to several billion dollars.
It’s not quite that simple, though, because this particular score is graded on a curve. The scores for every bank depend on where they land compared to other banks — there’s a set of denominators calculated every year.
This means, hypothetically, that if a bank bent the rules to lower its own score, its peers will look riskier in comparison. But it’s a different story if one country’s banks are all reducing their capital requirements in the same way, especially if they have the tacit approval of their local regulator.
That moves the focus to the Federal Reserve.
The TBIJ/ICIJ said the Fed would not clarify its position on banks netting longs against shorts. Their report quoted Senator Elizabeth Warren as saying: “I am deeply concerned that the Fed may be turning a blind eye as JPMorgan and other Wall Street banks cook their books and skim off funds meant to prevent a global economic collapse.”
In fairness to the Fed, there are a few reasons why they might’ve wanted to hand out a few mulligans on this one. They are significantly stricter than the Basel standard on a number of other issues, particularly year-end window dressing, and have come in for some pretty trenchant criticism from the industry for doing so. And the Europeans are not exactly risk-measurement angels. They successfully lobbied for a tweak to a different part of the scoring system which says “cross-border business” isn’t really cross-border if it’s within the Eurozone.
Even so, rule-bending would not contribute to an atmosphere of trust and co-operation among regulators, which seems important for their Endgame. The Basel Committee is making slightly pointed statements that “All GHOS members unanimously reaffirmed their commitment to implement Basel III in full”.
It isn’t clear that Fed was saying one thing in its published standard while doing another. But if it was, why would any other financial centre want to take the trouble of playing fair?
