Roubini: The Coming Global Stag-Deflation (Stagnation/Recession plus Deflation)

So should we worry that this financial crisis and its fiscal costs will eventually lead to higher inflation? The answer to this complex question is: likely not.
…
In conclusion, a sharp slack in goods, labor and commodity markets will lead to global deflationary trends over the next year. And the fiscal costs of bailing out borrowers and/or lenders/investors will not be inflationary as central banks will not be willing to incur the high costs of very high inflation as a way to reduce the real value of debt burdens of governments and distressed borrowers. The costs of rising expected and actual inflation will be much higher than the benefits of using the inflation/seignorage tax to pay for the fiscal costs of cleaning up the mess that this most severe financial crisis has created. As long – as likely – as these fiscal costs are financed with public debt rather than with a monetization of these deficits inflation will not be a problem either in the short run or over the medium run.

I am not yet convinced of the deflation thesis …

Posted by: b | Oct 26 2008 16:12 utc | 3

The simple fact is that your critique of CDSes is irrelevant to the vast majority of traded CDS agreements. The average leverage ratio of the S&P 500 is only 5, nowhere near the leverage of a Lehman (30) or Barclays (60). And most of the “55 trillion” in outstanding CDSes we keep hearing about has nothing to do with banks. Have you ever looked at the consituents of the itraxx index (Carrefour, E.ON, Vodafone etc??) How many of these companies operations are threatened by a higher CDS price? Almost none of them.
Moving a CDS price has no effective impact on most companies’ ability to conduct business. It doesn’t impact the company’s likelihood of default, and it certainly doesn’t impact the share price.
Furthermore, shorting a company’s publicly owned shares has nothing do with a company’s borrowing costs or its likelihood of default. Unless you’re talking about the very minor impact of convertible bonds, borrowing costs simply aren’t affected in any meaningful way by share price.

Posted by: vaudois | Oct 26 2008 22:37 utc | 5

You could also propose outlawing the “Enemization” of Islam, so that Israeli arms brokers remain the monopoly backdoor US arms sales to those “rogue states” through full knowledge US $30B grants over 10 years, everyone knows is a protected monopoly.
Certainly the stakes for humanity are just as high. Yeah, that’ll happen.
The bailout isn’t their money!! That’s all you need to know!! As AsstSecTreas Neel Kashkari said the other day under testimony* before the Senate Banking Committee, “The (bailout) money is to be used to pay down bad debts, not to hoard or used to pay dividends to investors or for leveraged buyouts”, then he paused,…
“but that’s more of a guideline.” (*- saw live, no transcript found)
You might suggest FDIC declare that bank robbers use of dynamite is null and void.

Posted by: Julia Compos | Oct 27 2008 5:30 utc | 7

There’s limited public info on credit default swap positions. market information is incomplete. But that cuts both ways. You can’t credibly claim that CDSes are poised to destroy the financial system without data, and the core complaint seems to be that there’s not enough data available. So this claim seems to be based on intuition. So is the intuition fair?
First of all, let’s try to agree that the “X trillion” statistic tossed around wrt CDSes is largely meaningless. The notional value of an insurance contract is by defintion many times larger than the referenced market – that’s just the nature of insurance contracts, *all* insurance of every type. The same is true of futures contracts. There’s 8 trillion USD in notional outstanding on the IMM Eurodollar futures contract alone, and some large multiple of this figure in options positions. Yet no one imagines there’s a crisis brewing there.
Second, let’s understand that CDSes, like all derivatives, are zero sum. That means that for every buyer, there’s a seller, and that every loss results in a gain for someone. And of course many positons are simply intermediated by dealers. CDSes are not assets, and no one is pretending they are. Their loss in value results in no loss in value to an entire system.
So let’s suppose that some people (let’s say, investment banks) own most of the unmatched CDSes, and that some class of sellers (let’s say hedge funds and reinsurers) has sold them. Is the risk that the hedge funds may default on their CDSes, causing problems for the banks? Is that the disaster that we’re trying to avoid by nullifying the CDSes? Because these to me sound like the same thing. For the law to nullify insurance on structured debt amounts to imposing a condition of uniform default, our own “worst case” scenario!
Maybe the idea is that hedge funds are *buying* CDSes and thus squeezing vulnerable overleveraged banks? This is what New York state is investigating. Never mind that hedge funds (small players relative to reinsurers) may not be able to manipulate this ‘gargantuan’ swap market. Forget for a second that borrowing costs are made up of much more than default risk, and that these changes in borrowing cost are denominated in basis points (hundredths of a percent.) Forget also that CDSes aren’t written only (or even mainly) on banks or banking issues, but on ordinary companies with sane leverage ratios (I keep referencing itraxx but for some reason no one seems to want to read it… oh well….) Is the idea that making bank debt uninsurable *by anyone* will make borrowing costs lower or the bond market safer? That’s a very weird idea.
http://www.nytimes.com/2008/10/23/business/23lehman.html?ref=business
5.2 billion changes hands over “globe threatening” lehman default.

Posted by: vaudois | Oct 27 2008 12:10 utc | 9

We currently have lots of insecurity in the financial markets which increases the perceived risk of lending and leads to higher interest rates. On could take away a lot of insecurity by dissolving them. That is one of Chua’s main argument.
That’s Chua’s statement. It’s only an argument if you provide reasons. Chua hasn’t offered any reason why abolishing an entire class of insurance – across dozens of unrelated markets – is likely to bring stability to any of them. He’s also not given us a reason why higher rates are ipso facto a bad thing: if I understand you correctly you disagree (and so do I.) As far as doing away with instability in markets — good luck. I’ll only note that research has shown that volatility in every studied asset class goes down with the introduction of traded derivatives (cf Mayhew & Mihov, Rao, Cavallo/Bologna et al)
Hmm – is the notional value of U.S. fire insurance higher than the value of the insured houses?
Of course it is, as the replacement cost of the house is always more than the market value. But as you also know insurance passes through dozens of hands before finding its way onto the books of a reinsurer in Bermuda or Zurich. Each time this happens the “notional amount” goes up by one, just as it does with futures. The notional amount of insurance written on US housing is at least 5 times the size of the total US housing market. Whereas the value at risk is many times smaller.
The Eurodollar crisis might be coming – with the large moves in currency markets right now, there is certainly the possibility of big problems.
The Eurodollar crisis might be coming. The world might end tomorrow. Or it might not. So far it hasn’t. But it might! The fact remains – no far no one, including yourself, has shown any problems to exist in the Eurodollar trading mechanism .
They are certainly not zero sum if the insured debt defaults
If the insured debt (an asset) defaults, that represents a capital loss, passed to the dealer through the mechanism of insurance. Capital losses cannot be magnified by the presence or absence of insurance that is zero sum. All derivatives are by definition zero sum, including CDSes. Nowhere in any market is the value of XYZ augmented or diminished by a side bet on XYZ. Simple arithmetic demands that the total amount “lost” via insurance can by definition be no larger than the capital loss experienced in the underlying market. If the asset is somehow overinsured, more money will simply pass across the table from a class of buyers to a class of sellers.
The whole industry is some 1.5 trillion big.
The biggest hedge fund in the world doesnt hold a candle to Swiss Re in terms of revenue, assets under management, or market power.
Debt insurance was very unusual only five years ago.
A most peculiar claim. Mortgage insurance has been around since the early 1900s, monoline insurance since the 1970s. Options on sovereign debt have been traded for decades. Debt insurance in some form or another has been around as long as debt itself.

Posted by: vaudois | Oct 27 2008 17:06 utc | 11

me – Hmm – is the notional value of U.S. fire insurance higher than the value of the insured houses?
vaudois – Of course it is, as the replacement cost of the house is always more than the market value. But as you also know insurance passes through dozens of hands before finding its way onto the books of a reinsurer in Bermuda or Zurich. Each time this happens the “notional amount” goes up by one, just as it does with futures. The notional amount of insurance written on US housing is at least 5 times the size of the total US housing market. Whereas the value at risk is many times smaller.
Wiki:

The notional amount (or notional principal amount or notional value) on a financial instrument is the nominal or face amount that is used to calculate payments made on that instrument. This amount generally does not change hands and is thus referred to as notional.

With that definition my statement is correct and your’s is wrong. The notional amount of five times re-insured insurance is one time the value of the insured object.
—
me – Hmm – is the notional value of U.S. fire insurance higher than the value of the insured houses?
v – Of course it is, as the replacement cost of the house is always more than the market value.
Is U.S. fire-insurance on replacement value? That would be astonishing as one then could “renovate by fire”.
Let’s see:

Guaranteed replacement value. Insurance companies used to pay out losses based on a guaranteed replacement value, which meant that a policyholder would recover the full cost to replace their home regardless of policy limits. Unfortunately, most insurance companies haven’t written this type of coverage since the early 1990’s.

Policies are written on actual cash value (ACV) of a house and there is only one policy for each house. That means that the total amount insured is equal to the total value of houses, not higher.
me – Debt insurance was very unusual only five years ago.
v – A most peculiar claim. Mortgage insurance has been around since the early 1900s, monoline insurance since the 1970s. Options on sovereign debt have been traded for decades. Debt insurance in some form or another has been around as long as debt itself.
You are right – I should have written exactly what I meant: Debt insurance via unregulated OTC CDS was very unusual only five years ago.
My other points stand.

Posted by: b | Oct 27 2008 18:25 utc | 13

it also occurs to me that Tavakoli’s own book on CDSes came out way back in 1998 (I’m a proud owner!)
My other points stand.
Obviously notional amount in the context we’re discussing (CDSes) has no finite referent, just like interest rate swaps. Is it really informative that there’s 38 trillion in interest rate swaps outstanding, even though the M1 is only 1.5 trillion? Why does the notional amount even matter, if it isn’t money at risk? In general a larger tradeable market would seem to offer more fairness and liquidity than a thinly traded market, wouldn’t it?

Posted by: vaudois | Oct 27 2008 19:27 utc | 15

vaudois – to stop the silliness – tell me two simple things:
1. where is the social need and benefit of CDS as mankind has obviously survived for 10,000 years without them and we only introduced the first of them 10 years or so ago.
2. what are the social costs of CDS and the credit structures allowed by using CDS that are currently seen as obviously infecting the real economy.
Does the need/benefit justifies the costs?
I say no. You might benefit from them. The society, in my view, does not.
The issue is the same as with Enron (and why I declined to work for them). The Enron scheme made profits for few at social losses for many. That they created a market for CDS is not really a positive argument in my view.
As CDS imposes risk on the real economy and society while only create profit for a few they should be abolished.
The best way to do so is by declaring them null and void in a controlled process. I’d rather do that quickly while others may argue that this can be extended or that we can create some exchange etc.
As if exchanges never get blown up or markets, no matter how transparent, crash once a while.
Where is the benefit for society for allowing CDS? Oh – cheap rates?
Yes sure – you could buy a house in a NINJA loan with 0% down at 1.5% starter rate because there where CDS that insured the CDOs that were chopped up liar mortgages. Did that really help the society?
We should get rid of them and should do it fast.

Posted by: b | Oct 27 2008 19:46 utc | 17

1. where is the social need and benefit of CDS as mankind has obviously survived for 10,000 years without them
The “social need” is debatable, but the social benefit (as I see it) is the same as the benefit conferred by IR, forex and any other derivative: stripping Wall Street of its monopoly over capital markets. Permitting as broad a market as possible the means to bid and offer risk capital on even terms.
Having traded these things for 15 or so years, I view derivatives markets (especially in liquid, exchange traded instruments) as the most democratic markets in the world, as close to an idealized form of capitalism as you can get – transparent, fair, and efficient – exactly what the world of fixed income was not before futures introduced competition, but where it will likely return if society chose to ban them.
You could plausibly claim that CDS markets aren’t these things (and I’d agree!) but that’s just an argument for exchange clearing and common standards. Under the right condition CD swaps would ultimately be a much more democratic medium for valuing and transferring credit risk than the corporate bond market, especially for those like us with no inside information of a company’s balance sheet.
i’d add that the “risk on the real economy” is only perceived because of some misguided, unecessary and undemocratic plans to subsidize investment banks bad risk- and cash- management.

Posted by: vaudois | Oct 27 2008 20:54 utc | 19

Let’s not ban CDS. Let’s simply require CDS seller capital requirements (no dynamic hedging! no reliance on “off-setting” CDS!) at levels that would protect against a systemic crisis. That would pretty much kill the business model.
Too cheap risk insurance (CDS) was a huge factor in this crisis. Force it to be properly priced.

Posted by: jeff65 | Oct 27 2008 22:47 utc | 21

Seems like am not the only one who sees problems with these shiny objects that are called CDS. Did someone upthread say notional values don’t matter?
A longer piece in The Institutional Risk Analyst: In the Fog of Volatility, the Notional Becomes Payable
Just the conclusion:

In both cases, the normal operation of the OTC derivatives markets is creating a cash position that must be funded in the real world and is thus distorting these benchmark cash markets such as LIBOR. This distortion is magnified by the dearth of liquidity due to the breakdown in the rules regarding valuation and price. So far, the Fed and other central banks have addressed the on-balance sheet liquidity needs of global banks. But as retail and corporate default rates rise, funding the trillions of dollars in notional off-balance sheet speculative positions in CDS, which become very real and require funding when a default occurs, could prolong the economic crisis and siphon resources away from the real economy.
…
Macroeconomics models are highly evolved hypersensitive systems with somewhat linear characteristics – at least until they are driven into instability by discontinuities in underlying ground rule and assumption sets. They are all fed by statistically smoothed inputs; the same kinds of statistical methods that people used to turn collections of insufficiently documented Alt-A mortgages into allegedly AAA rated pools and securitizations. Present market conditions have further eroded the clarity of these inputs. In math terms, the widened bands of dispersion have increased uncertainty with regards to locating the true center of the input sample set. This reminds us of the old Cold War SIOP (Single Integrated Operational Plan) models where you could change the outcome of the war by arbitrarily tweaking one variable 1/10th of a percent.
To us, a big part of the way back to restoring confidence in global markets is for regulators and shareholders to start redefining the types of activities and products that are reasonable for global financial institutions and under what capital requirements. The open ended, anything goes environment of the past decade has destroyed public trust in financial markets, thus the palpable fear visible from street level banking customers to the largest institutional players. Until analysts and investors begin to reach a consensus about value and price, primarily by setting clear rules regarding market transparency and disclosure, the ability of markets and regulators to assess or measure risk is seriously compromised.
But until global regulators and politicians summon the courage to restore reasonable rules to the financial markets, particularly a balance between risk allowed and transparency required, attempts to understand price and value, and thus capital, will be fruitless and the confusion will continue. Markets like societies, after all, must have clear rules in order to function properly. What a shame that private market participants do not have sufficient political wit and savvy to lead the discussion and instead must await the judgment of the same political class that caused this mess in the first place.

Posted by: b | Oct 29 2008 15:19 utc | 23