Monday, April 7, 2008

Profits? We Don't Need No Stinkin' Profits!

One question I like to ask frequently without ever receiving a satisfactory answer is the following: Where will future profits in the banking industry come from? Forget about all the write downs, despite the fact that they still aren't completely behind us, why should I buy any financial stocks now? Global debt issuance has plunged is all areas of the debt markets. According to the Financial Times, total debt issuance volumes were down 48% from a year ago. Down $1 TRILLION. That's with a capital T! Total syndicated loan volumes were down 47%. Structured finance was down 89% (ouch.) Debt underwriting was a large profit center for banks and brokers and that was just cut in half. With the Fed's frantic efforts to bail out the banks, one can make the argument that these markets will improve. But will they get back to prior levels? I don't know. Ask investors who are holding AAA CDO's trading at pennies on the dollar if they're ever going to buy another CDO again?

Meanwhile, speaking of $1 trillion dollars, primary dealer fails surged to $1 trillion for the week ending march 26th, up $804 billion from the prior week. What does this actually mean? Let's assume that all the fails happened to one dealer (Dealer X.) Also assume that Dealer X has $1 trillion in inventory owned at an average rate of 5%, all treasuries, that he finances every day in the repo market which ordinarily trades around fed funds. The dollar value of a basis point on $1 million overnight is $.28. So every day, the dealer earns the difference between the interest he collects on his inventory ($1 trillion x 500 basis points x $.28 x 1 day=$140 million) = less what he pays to finance the inventory ($1 trillion x 225 basis points x $.28 x 1 day = $63 million.) Assuming no mark to market changes in his inventory, he should collect $77 million a day in carry. This is why banks love a steep yield curve. They can buy long-dated securities at a high rate, and finance them at a lower rate short term in the repo market, using very little capital. However, if a dealer fails to deliver collateral into a repo (which requires delivery versus payment), he is still obligated to pay the interest, even though he never actually gets the loan. This leaves the dealer no choice other than to take out another loan to finance the securities that he failed to deliver, thus essentially doubling his financing costs. Assuming Dealer X failed on his entire inventory, his financing charges just doubled reducing his profit in the above example to $14 million a day. This example is merely meant to explain the mathematics of fails and how excessive fails can cost dealers significant amounts of money. Furthermore, it is meant to show how great the dislocations are in the financing markets and that the Fed absolutely needed to intervene to swap treasuries for MBS to help alleviate some of the strains.

A closer look at the Fed report shows that the difference between reverse repos and repos for mortages, agencies, and corporates is approximately $1 trillion (total reverse repos for MBS, agency and corporates = $600 billion while repos = $1.6 trillion). So if dealers attempted to reduce their financing needs by not renewing any of their reverse repos in non-treasury securities, and merely focused on financing their inventory, they would be left with $1 trillion dollars to finance. That's with a capital T. And that's why dealer financing at the discount window has soared to $38 billion. Add this to the hundreds of billions lent by the Fed through the TAF and TSLF and the $30 billion loan to JP Morgan to buy Bear. Clearly, dealers really need the Fed, as other sources of financing for the $1 trillion in inventory has disappeared. The Fed is cooperating, which is good. Is it enough? If it were, would the fed funds rate have had an 8.5% trading range on 4/4/08, topping out at 10%? This debacle will only be over in my book, when dealers can get financed themselves without a crutch from the Fed. In the meantime, the world will continue to unwind its loser bets from the glory days of 2005-2007. And profits at dealers will suffer because they have nowhere to turn to sell their products. Can they make it all up in carry from the Fed's easy monetary policy? It's what the bulls think and it's why they all think the worst is behind us. But I'd bet against it.


Oscar said...

Two questions about the repo fails:
1. Why would an institution fail? If it fails to deliver once, why would it not fail a second time and a third? Is it a symptom of negligence or stupidity? (That's sort of what I read into your comment, btw, K10.)
2. What institution are on the other side of repo lending? Other financials? Won't, then, their profits be higher than they'd expected if they are getting windfalls of repo penalties?
That seems like a zero-sum game.

K10 said...

Dealers fail to deliver for a few of reasons. 1.) operational problems which they don't notice until it is too late to fix. 2.) trader error-the trader either did a trade that the counterparty disagrees with or just inputed the ticket incorrectly. 3.) they failed to receive securities on the other end. Although errors are sometimes caught at the end of the day, often it is just before the wire closes and there is not enough time to fix the problem. Obviously if the dealer failed to receive on one end and then fails to deliver, it is not a cost to the dealer. However, if you are waiting to receive $2 million of a t-bill in order to deliver a $50 million block, and you end up failing on the $50 million. So usually at the end of the day, there is a scramble to borrow the $2 million just to be sure you don't fail on the $50. The business of "picking up fails" is a big part of the job of being a repo trader, as one fail can wipe out profits on the entire portfolio for the day if you aren't careful.
Institutions on the other side of repo transactions are central banks, (huge buyers of treasuries and agency debt) mortgage reits, hedge funds, money market funds, and probably other institutions I've forgotten about since the 90's.

Oscar said...

OK, but it seems that a spike in fails isn't necessarily a symptom of a slowing economy or problems within the financial sector, right? I can see halved debt issuance as being a big red flag. I'm still unconvinced that I should be reading anything into the fail data, though.
Anyway, thanks for replying.

K10 said...

oscar, you are right. it is not a direct sign of a slowing economy. a spike in treasury fails is merely an indication of the enormous amount of hoarding of treasury issues. hoarding treasuries is a sign of panic in the money markets. if the money markets can't function properly then dealers can't finance their inventories and are forced into reducing positions and delevering further. if dealers are continuing to delever, they become net sellers of securities, meaning there is less investment and less lending. that is what ultimately effects the economy.