This is the second time in a week that we're falling all over ourselves wondering why swaps are through Ts. It's not hard: IRS is collateralized, Treasuries are not. With IRS every 3 (or 6) months, you have to cough LIBOR in order to receive the fixed leg. So essentially your maturity commitment is never greater than that, and indeed, for the vast majority of swaps receivers, mark-to-market collateral adjustments happen daily, so your credit risk is 24 hours! This is not the case for USTs where if you invest for 10 years, you have no clue for every single one of those 10 years if your principal will be repaid, and there is absolutely zero collateral adjustment. Why is this so hard for people to understand?
Ts are being printed without even the slightest regard for the (old-fashioned) concept known as fiscal rectitude. Swaps can be printed all day long but if the rates start moving you get paid/have-to-pay every 24-hours for the assumed risk. It follows that Ts are now a much riskier bet than swaps, and indeed, what surprises me, is why it has taken so long for this to be happening.
Nobody (to a second order approximation) knows what the fuck any of that means, because, since they are sane people with lives to lead, they don't delve into the details of advanced finance.
let me make it perfectly clear. Lend money for 24 hours is less risky than lend money for 10 years. Does that make sense?
On your point about finance being complex. This is Bloomberg writing the story. They of all people should know. And this is the second time in a week that they're pushing this stuff.
BTW: for the average finance guy the CS jargon on this site would also qualify as something to be dismissed by "sane people with lives to lead". Not a valid argument. And before you tell me that this is a CS site so that is understandable, I'll ask you why this Bloomberg stuff is making it to Page 1.
If there's any chance of a treasury default or other disruption, what happens the next day when nobody wants to be left holding the bag? Seems like in a scenario where something is up, daily liquidity is more dangerous.
there is indeed an argument that says that daily mark-to-market (with collateral reconciliation) nullifies all but the very wide tails of the distribution. So the 6-sigma risk is paradoxically increased on daily liquidity, because you're sleepwalking into taking bigger risks, on which big sigmas are never materially manifested......until they are...
Basically - we're erasing "normal" counterparty risk in derivatives, in so doing blinding us to big abnormalities. Imagine an ocean with zero waves (ie a pond), until the tidal wave hits. Wouldn't it be better to have some fear of waves hitting you daily, so you're (somewhat more) prepared for "the big one"?
Scarily, there are something like 70 trillion of swap contracts outstanding. Someone correct me if I am wrong (though I'd bet I'm underestimating).
There is a large and growing body of evidence that huge central bank involvement in markets is increasing exactly such risks and it is an extremely important point.
Still... if we consider that you wear the up-to-6 sigma risk with Treasuries (ie 99.999999% of the time - whatever the number of 9s is), but you are spared all but the 6-sigma risks in swaps, then it is probable (though I agree .. not certain), that you're better off with the mark-to-market collateral protection that you get in swaps. Thus the article's main point is erroneous.
It is clear that the idea pushed by Bloomberg here, that banks have somehow become better credit quality than the US government, is wrong, as can be seen by a simple look at their equivalent-maturity borrowing rates in the bond market (well above Treasuries). IE: not comparing apples with oranges.
> Ts are being printed without even the slightest regard for the (old-fashioned) concept known as fiscal rectitude.
> Scarily, there are something like 70 trillion of swap contracts outstanding.
I guess there's 15-20T in T's outstanding... if there's 70T in swaps (that don't have a Fed or an army behind them), what would your comment on their 'fiscal rectitude' be?
Here is the perfect opportunity to clarify a misconception.
Swaps != debt.
Swaps do not create a liability/asset unless interest rates move, and even so, most of the swap contracts will then nullify one another.
Most banks admittedly have trillions of notional swaps on their books. The very vast majority of these offset each other. The "open position" (net exposure) in banks is very heavily monitored by the authorities. Banks are not heavily exposed.
Arguably, corporates (who hedge) and insurance companies or pension funds DO have open positions in swaps. But even so, recall that the potential liability is proportional to the likelihood of an interest rate move. Let's say an open position of 1 trillion (enormous) receiver swaps (lent) sees a 3% upmove in 5y interest rates on an average 5y maturity of the book (usually lower). Assuming rates near zero which they are, the duration calculation means a net debt becomes payable by such company of 3x5% = 15% of the notional = 150 billion USD. So 1 trillion became 150 billion, even under an extreme scenario. 1/6th. Then recall, that rates will never move 3% (300bps) in one line. They'll move a few bps every day. It'll be very clear once the liability hits, say 5-10 billion, that the counterparty cannot pay anymore, they will be made bankrupt (or close these positions) well before 3% (remember daily margining?), and the swap contracts will become null and void immediately. So now 1 trillion of swap contracts became 5-10 billion realizable debts.
By contrast, when the Treasury borrows 1 trillion, they owe a full trillion, no matter what happens in interest rates. See the difference?
Now in this analysis, I'm glossing over counterparty risk in swaps. Banks "not heavily exposed" relies on the idea that all banks are good for their commitments, since their netting is done against other banks. This is not the case in a Lehman style scenario(it's like a huge web that even neo4j would have trouble monitoring - ha ha - if a piece of the web falls away, many banks are then potentially exposed). Hence the major focus of all policy makers in the past several years has been to make derivative clearing daily-marginable, and exchange cleared. To erase that problem (outside of 6-sigma "impossible" etc - see one of my other posts)
Further to that policy maker regulation, did you know that in the latest European banking sector rules, derivative contract commitments are senior even to small depositors? So equity goes first, then subordinated (higher yield) debt, then depositors > 100k, then senior debt, then depositors < 100k, and only then, swap liabilities.
So basically what you're saying is that swaps under these new rules are good as they go south quicker and more obviously in changing market conditions, which forces the hand of the holder to plan ahead and actively hedge their bets elsewhere in real time rather than relying on the long term false security of something that might be rotten underneath?
Well, swaps will not move south any faster or slower than treasuries, usually, but the former will be a leveraged instrument so will tend to force the holder's hand to close much more quickly.
Generally I did not intend my post to exonerate swaps as a dangerous instrument. They are. Just to say that they cannot be equated with outright debt.
Nice explanation in the above post... the kind John Merriwether (LTCM) might have given in 1997 or Dick Fuld (Lehman) might have given in 2007. You're certainly right in that there's less than 70T in leverage involved in the swaps, but I think you're also right to call them dangerous.
Tomorrow a buck is worth about a buck. Given some nice 70s era stagflation, in a decade a buck might only be worth fifty cents. Loaning one dude a buck and getting fifty cents back isn't all that different than loaning out ten dimes and getting only five paid back.
There is also downside risk. Your financial life will be about the same tomorrow as it is today, so I can reasonably estimate the odds of getting a buck back if you need change for the vending machine or whatever. Ten years from now, well, hard to say how many bad things can happen to someone in ten years.
Agreed. But lending to a riskier counterparty for 24 hours is usually going to command a lower risk premium than lending to anybody, even the US government, for 10 years. That's the essence of the story here. You make a valid point though: until now it has been riskier even to lend for 24 hours to a bank because of the binary nature in which banks go bust: they're solvent until they're not, and the run can happen in an event-wise, unpredictable fashion.
The real story then is that even taking into account this high banking-sector risk, the US government is now riskier. But this has much more to do with the unlimited printing of debt (the real story) than the IRS market swap spread, and my wider point is that daily margining has been ignored by many of the people (here and in the financial markets chatter) who are talking up this issue as some kind of rubicon moment. This has been in the making for years and the slow grind below zero is not something that should be seen as extraordinary, nor, given the aforementioned slow grind lower, should it come as any surprise that it can go negative.
> But this has much more to do with the unlimited printing of debt (the real story) than the IRS market swap spread
I don't get how U.S. debt is the driving factor. The rate of federal net debt growth has actually been slowing over the past ~3 years as the federal deficit has shrunk, i.e. the arrow is pointed in the right direction.
He's making the case that the simple fact that a Treasury can have a 10-year term makes it riskier than a swap that is marked to market every 24 hours.
This might be a makeable case, but just comparing the terms is not enough to do so. And, while he is using it to explain why Treasuries seem to have a risk premium now, it doesn't explain why they did not previously. Did swaps just start being m2M every 24 hours? Certainly there has been no major change to the nature of Treasuries over the past few months. So why the sudden shift?
vegabook, I encourage you to make a non-hostile comment explaining in plain english how swaps are much closer to 24 hour loans than the underlying securities. This is a good explanation and deserves to be near the top of the comments on this article.
The jargon and tone of your original comment is the reason it is being down voted.
apologies. It's not obvious because the jargon is opaque (though I still question why, if it's not understood, it's upvoted to page 1).
In swaps, as the word suggests, you must pay (the prevailing short-term interest rate) every few months for the right to receive a pre-agreed (long term) interest rate. The key is that you must pay, and if your counterparty does not pay, you will not pay either. So you have a much lower credit risk than an outright lending of money since if the counterparty starts failing, you can stop paying. Sure there is still some credit risk, but it's nowhere near as high as it would be to lend money for 10 years "naked". And since the crisis, moreover, we have "daily margining", that is, if you win in any 24 hour period, your counterparty must pay your winnings. If he doesn't you can cancel the contract. That's much lower risk than betting on repayment with no feedback, for 10 years.
I'm just generally quite "surprised" that this stuff is coming out of Bloomberg who should know better.
I think perhaps you aren't looking at this clearly, because with a swap you have exposure to forward credit spread premium - the market's implied expectations about where LIBOR will be versus the rate from the near risk free overnight rate curve. In 2008 this blew out, and in spite of institutional changes, it's still possible that one day this might happen again.
So you do in fact have a long term exposure to forward credit quality, just as you do with UST, except that the clearing house has no taxing power and cannot print money.
The primary deficit is shrinking, whatever one may think of Mr Obama - and that's hardly surprising when you look at unemployment. So your claim about causality requires further articulation.
It's also not the case that a UST position is not collateralised. If I sell short 100 mm tens, and I borrow these on the market against cash collateral from my receipt from the sale then if the price drops 10 points I get that back in my bank account - just as with a cleared swap position.
Ts are being printed without even the slightest regard for the (old-fashioned) concept known as fiscal rectitude. Swaps can be printed all day long but if the rates start moving you get paid/have-to-pay every 24-hours for the assumed risk. It follows that Ts are now a much riskier bet than swaps, and indeed, what surprises me, is why it has taken so long for this to be happening.