Armageddon Part One is over. The question, like a hurricane, is whether the US is now in the eye of the storm or is it clear sailing from here on? The "navigators" are mumbling something about "all clear", but then that's what they mumbled last time.
Follow the money… how much money got "pumped" into the US economy after the start of the credit crunch compared to before? It's a little confusing:
a. According to his testimony to the House Financial Services Committee on 21st July by Mark Zandi (.pdf) the head of Moody's (NYSE:MCO) - so far the Fed has pumped $2.7 trillion into the "legacy" banks and Congress, via the Treasury, pumped $1 trillion directly into the veins of the economy via the stimulus packages. Presumably those are the correct numbers; certainly no one jumped up and said he had his arithmetic wrong.
b. The testimony also has a summary timeline of issuance of securitized bonds, (source Thompson Reuters).
c. A recent report from Thompson Reuters gives a chart of debt created by the "legacy" banking system, of the traditional "Originate to Hold" variety.
d. Then there is the increase of National Debt, which is easy (although different sources have slightly different numbers), presumably that was achieved by sales of Treasuries (I gave up trying to figure that out from the Treasury Website).
e. As a check there are the estimates of the total debt in USA which I referenced from Morgan Stanley. Theoretically (b+c+d) should add up to (e), it does more or less; I imagine any discrepancies are due to roll-overs which I didn't find data about.
Put all together it tells a story:
Big picture: what appears to have happened is that the Treasury has taken over the heavy lifting that securitization used to do in terms of providing the "credit-driver" of the US economy. That is not healthy, and it may turn out to be unsustainable.
Looking at more or less the same data another way, this is an estimate of what happened in the eighteen months from January 2008 and the previous eighteen months:
I'm a bit confused by what looks like a $700 billion increase in High Grade Corporate Bond Issuance, but the data doesn't account for rolling-over so the net might be less.
Be that as it may; the important point is that the $2.7 trillion of "bail-out" money doesn't get into the economy until the banks start to lend it. So far they are not doing that. They say that's because no one wants to borrow; the borrowers say that's because the terms offered by the banks amount to legalized loan sharking.
Whatever, it's not getting lent.
By that logic, the amount of credit plus "one-off-stimulus" that got injected into the main artery of the economy is 22% down on the preceding eighteen-month period. Taking into account that a proportion of the stimulus packages didn't actually get shelled out so far, add in an allowance for rolling-over, the actual decline year on year could well be in excess of 30%.
That's not going to produce hyperinflation in a hurry.
Rather the opposite, or perhaps what the head of the EU Central Banks sensitively calls, "disinflation"; whatever word you use, that's what appears to be happening. Although no one except mavericks like Mike Shedlock are uttering the "D-Word".
That may well be a "good thing", over the past eighteen months there has been a growing consensus that the USA had much too much debt and that a bit of "deleveraging" might not be a bad idea for a change.
But regardless of whether that's right or wrong, that is NOT inflationary. And it's not going to create either jobs or economic growth, nor is it going to get the traditional engine of world economic growth, the US consumer, to go out and max their credit cards.
It's all very well that the Fed is dropping the base-rate to zero, but that's not getting passed on down the line. If you got a good idea in the USA right now and you go looking for credit, at that "risky" end of the spectrum you will pay 10% to 12% over LIBOR. That's hardly a "stimulus"; particularly since there are places in the world where you can find ways to pay a lot less.
That's what happened in Japan after their housing bubble, the zombie banks got saved and showered with cheap money. Everyone else paid through the nose, and the economy tanked.
Without securitization, that could happen in the USA too.
It's people with good ideas that create "real" new economic growth. Building underpasses for turtles or tagging sage grouse (which is the type of "shovel-ready" project the stimulus plan is financing) might be "nice", but that doesn't create sustained economic growth.
At the current marginal productivity of that kind of credit, $1 trillion of new debt generates about $300 billion of extra GDP, every year, theoretically forever and ever, and that generates about $100 billion of tax revenue, every year.
By contrast, $1 trillion of "stimulus" generates (presumably) about the same amount of GDP once, like a paper fire. After that it generates nothing because the turtles don't pay to use the underpass, except of course that someone has to pay the interest (about $30 billion a year) and someone (like you) also has to pay back the principal at some stage.
Discount all that and the "credit-engine" of growth now in the USA is about half what it was in 2006/7.
Sure, perhaps the Fed and Congress will pump in more (they have pledged another $8 trillion and that's not counting the "unlimited" line items), but it will have to be a lot more than what's going in now to run even the remotest risk of inflation, like about $2 trillion by the end of the year and $3 trillion next year. There is no sign of that happening. Particularly since the ultimate source of that will need to be from selling US Treasuries, and there appears to be a limit on how many of those you can sell in a year, except to the Fed.
Traditionally (in recent years), they shifted about $500 billion a year. The target in 2009 is $2 trillion of which about $1 trillion got sold, but there are signs that the second trillion is going to be a lot harder to shift than the first.
Of course, there is always "Quantitative Easing". That's when the Fed prints money without posting collateral with the Federal Reserve Agent, but there's a limit to how much that can happen, or even how much it can get away with.
One thing I don't understand (perhaps someone can explain this to me in words of one syllable please), how come the Fed's assets "only" went up by about $1 trillion (if you can call toxic assets and IOU's from bust banks, "assets").
But they say they shoveled $2.7 trillion into the banks plus another $300 billion or so for Currency Swap Lines? Maybe I'm simple minded but isn't there about $2 trillion "missing"? Perhaps Chairman Bernanke's interrogators were asking the wrong question, like not so much "where did it go" as "where did it come from?"
Be that as it may, until that money gets lent out (which it isn't), even if he was pulling a fast one, I guess that's basically just another way to say "forbearance".
Options going forwards:
It looks like there are five choices (quantitative easing on any scale is not a serious option):
(a): Persuade America to get by on half the credit that it has gotten used to (let unemployment rise to 15% and live with a declining nominal GDP) until that comes down to a more manageable level. That's what the IMF would insist on if the USA was a Third World country that got itself into this sort of a jam.
(b): Massively increase the National Debt by selling $2 trillion to $3 trillion of Treasuries a year, year-on-year (remember that has to finance the budget deficit and the current account deficit too). There are two problems with that, the most relevant is that it's highly unlikely that in the current circumstances the USA will be able to sell more than about $1.5 trillion year.
(c): Massively increase taxation and fire a lot of government employees. That would be popular.
(d): Get the "legacy" banking system to ramp up lending. The way to do that would be to nationalize them, like they do in China. That wouldn't be hard, most of the big banks are more-or-less owned by the government anyway so that is certainly a viable option.
(e): Fix securitization.
Well (b) is basically impossible so that's not really an option, (a) and (c) could result in riots and (d) might precipitate a fascist coup d'etat. So that appears to leave just (e) as the only viable option, apart from muddling through, with the emphasis on muddle.
So why is no one doing anything about that?
I went to a conference in London recently that grandly proclaimed was going to map out the Great-Leap-Forward that would Save-The-World as we know it, namely by re-starting securitization.
The impression I got was that no one had a clue how to do that. You have to remember that securitization as we know and love it today was a product of thirty years of financial "innovation". It's hard to imagine that anyone can sit down for five minutes, figure out what went wrong, fix it, and then the global financial system can live happily ever after.
In the conference there was a lot of talk about what NOT to do the next time around, but nothing remotely coherent about WHAT to do. Someone from of the EU Banking Committee helpfully clarified things by saying, "we have many urgent matters to deal with; securitization is at the bottom of the list". That was about the clearest direction that anyone came up with at the conference.
From the moment the "crisis" started in earnest in about July 2008 when Secretary Paulson proclaimed "The US Banking System Is Safe And Sound", the reaction of the market to that information was to short Fannie (FNM) and Freddie (FRE) into oblivion, the issue of securitization was given a wide berth, like a truck full of rotten fish.
Secretary Paulson had a peek with TARP then retreated from the stench. It was the same story more or less with PPIP. The only effective player was the Fed via the TALF program, but the effects of that were muted.
Mark Zandi treated the specter of that dreaded subject with kid gloves in his testimony, he said; "It would be a mistake to scrap securitisation altogether".
Which means what? That the idea of scrapping securitization altogether is on the table? Perhaps that's not surprising; after all Moody's were supposedly the bad guys, and well "securitization (the core of Moody's business model), caused the credit crunch, right?" So best just to let that dying dog lie until the storm passes over?
In the same vein the emphasis in Secretary Geithner's eighty-five page Financial Stability Plan puts the priority for fixing securitization somewhere below controlling executive compensation and relying less on ratings agencies.
That's hardly a "Plan". And in any case whatever hazy ideas it puts forward on the subject (and they are truly hazy), they will have to wait until the whole package gets passed into law, which at the present rate of progress might be Christmas 2010.
In essence, the plan is very similar in structure to the plan to build the Maginot Line translated from the original French with one or two words changed. The emphasis is on preventing what happened before happening again, and having the tools to deal with the unexpected next time something unexpected happens rather than something "expected", if you follow the logic?
There is reassuringly a lot of talk about "firefighting". That was the theme of Secretary Geithner's first presentation to the Senate Financial Services Committee, where in response to a question from Senator Dodd he said "you can't fight a fire with a committee".
Quite right too, but I'm just wondering, where's the fire?
In case no one noticed, the problem of starting fires doesn't typically arise after you just put one out and the burnt out shell of the building is inundated with water. And in case no one noticed either, the American economy is on its knees; forget about fires, the danger right now is drowning. Perhaps the Great-Leap-Forwards to build a Fire Department can wait, certainly until there is the remotest chance that someone might be able to start a fire?
The important point is from 2000 to 2007 about $14 trillion of securitized debt was created and sold, about half of that was sold to foreigners. By contrast, $4.5 trillion of Treasuries were sold (40% to foreigners). The securitization pipeline to the debt used to lubricate the US economy AND to foreign exchange to help finance the endemic current account deficit, is now blocked. It is highly unlikely that it can be substituted by selling Treasuries.
If either securitization is not fixed or an alternative is not found, some hard "readjustments" may be in store, Armageddon Part Two perhaps?
How can securitization be fixed and QUICKLY?
The best (and only) coherent idea I saw so far, and in theory it looks like a good one, came from Professor Perry Merling of Columbia University. If you want to read a clear and well written analysis of the credit crunch you don't need to look any further than here (.pdf).
Merling's idea is that the Fed should become the insurer of last resort as well as the lender of last resort. The logic there is that what's missing out of the equation at the moment is that no one wants to sell insurance on the possibility of a toxic asset defaulting (i.e. a Credit Default Swap (CDS)), so the Fed should sell insurance (at a high price).
That's because (as a rule of thumb) the value of a toxic asset is equal to the value of an equivalent risk-free-security (i.e. a Treasury perhaps), minus the cost of insuring the toxic asset. The problem right now is that since the cost of insuring a toxic asset is approaching the amount insured, the nominal value of all the toxic assets sitting on the banks' balance sheet (or off them - same difference), is approaching zero by that rule of thumb.
That makes perfect sense as a short-term fix, although the practicality of that idea is a bit dubious. Even if you are selling insurance at a high price, you still need to know something about the insurance business, which in practice is slightly more complicated than just being the lender of last resort (all you need to do there is keep an account of what you lent and make sure that it comes back with interest, and if not you go in and repossess the executive jet or something equally unfriendly).
And the fact that the Fed owns AIG doesn't exactly provide an ironclad way to load up on that expertise, since it was AIG selling insurance much too cheap (about 20% of what they should have sold it at), that created the problem in the first place.
Also, and perhaps I'm being thick again, but I don't understand why the Fed should be the insurer of last resort when there is a perfectly good insurer of last resort that has been in business for over three-hundred years and so far always paid up on its claims (more or less)? It's called Lloyds of London (and I don't imagine they would write any insurance to anyone who does not have an insurable interest, which might be a good thing too).
But in any case, the Treasury appears to be having a go at sorting out the CDS "market". Why am I not convinced that this won't just create more confusion and systemic risk?
How about if instead of an 115 page document full of hazy ideas and multiple potentially conflicting layers of supervision, there was a one line mandate along the lines of "All CDS must be re-insured (not hedged) by a competent and approved re-insurer, and not less than "X%"", and let the market sort out the rest?
Two other ideas:
1: Covered Bonds.
The problem at the moment is that no one trusts the valuations of the toxic debt, so anyone who holds the stuff is hanging on to it because they believe it is ultimately going to be worth more than what anyone (else) is prepared to pay for it now.
That will presumably sort itself out at some point, and now that the Fed and the regulators are practicing "forbearance" there is no particular hurry.
But the big problem is that no one wants to write any more of that stuff because they know they won't be able to sell it at a half-way decent price.
That is a problem. There are good quality borrowers in America (a few of them left), that are crying out for debt at the sort of cost that securitization could deliver (when it was working). In the normal run of things, people need to roll over debt (and there is a lot due for rolling over in 2011 to 2014), municipalities need to finance infrastructure, and people with good ideas need debt to put them into practice. The legacy banking system of originate and hold does not have the capacity to service that demand.
So how about forgetting about the old way of doing securitization put in place, or mandate a method of doing securitization that everyone trusts.
There is one. It's called covered bonds, where the liability of default rests with the issuer. That type of securitization was invented in Germany two-hundred years ago, and its credit history is impeccable. They are slightly more expensive than the way securitization used to be done in the USA. More importantly, the arranging banks make less (which is probably the main reason they never caught on in the USA).
That idea has been wafted around like a "well, perhaps that would be nice". Mark Zandi mentioned it in his testimony, others have had that idea.
Well perhaps it would be "nice" if until someone comes up with a better idea, the government mandated that the only securitization of pooled mortgages in the USA would have to be via covered bonds, using a proscribed structure laid down by law, and said that like...tomorrow.
Writing that law wouldn't be hard, all that would have to happen is to get a copy of the German Law and translate it, that would at least get some securitization going.
What is a really bad idea is for the government to have a go at inventing something new or fixing the thing on the hoof.
First, that will take too long, second it is the job of government to lay down clear practical regulations rather than a Pandora's Box of hazy ideas in an attempt to do what the market got wrong.
Sure the "financial innovators" screwed up, but that most certainly does not mean that the government can do better. The risk of failure will be that a good portion of the US needs for credit will have to be "manufactured" via Treasuries, which would not be healthy. That's loony Socialism, the next step is deciding who gets what, like, "are turtles more important than energy?"
Let the industry sort out the mess, but they need breathing space, like two or three years, and covered bonds can provide that.
Everyone talks about "transparency" these days, like they used to talk about "free love" at Woodstock. What does it mean?
What it means is that market participants need to be provided with sufficient information about the stuff they are buying in the marketplace to be able to make rational and well informed decisions.
The problem now (and it was then), is that market participants are not provided with sufficient information to make rational and informed decisions.
And they did not. The proof? They paid too much for toxic assets; and that had a feedback effect fueling the bubble because people manufacturing those "assets" found they could make a fortune packaging them up and selling them like melanin tainted milk.
There were a number of reasons for that:
(a): People trusted the ratings and they did not check them (and they should have, the ratings simply defined whether or not you could consider them as investment grade in front of the regulator).
(b): The insurance against default was sold too cheap.
(c): Although there was a charade of marking to market, in fact that was done either by reference to about twenty benchmark toxic assets trading on the ABX Index (like valuing a goat by finding out how much a cow is selling for), or by the "rule of thumb" that the value was equal to the value of a Treasury minus the cost to buy a CDS, which were traded on a supposedly liquid market.
That's not the right way to do a valuation.
What is most extraordinary is that was the way that about $20 trillion of mortgaged and asset backed securities got valued by market participants, which flies 180 degrees in the face of sensible or coherent valuation methodology. Proof of that pudding is that rule of thumb manifestly did not provide a reliable measure of value; in a nutshell that is Voodoo Valuation Standards.
That the valuations were wrong is self-evident, that's what the credit crunch is all about; the bankers found out that they were off the mark by $2.5 trillion to $5.0 trillion (plus or minus depending on who you talk to and including the rubbish that was fobbed off on foreigners, i.e. about half).
And that lunacy is what triggered the catastrophe, when the price of CDS went sky-high. The so called "mark-to-market" Alice in Wonderland valuation methodology marked all the toxic assets down to zero. Talk about Dumb and Dumber.
The right way to do a valuation (in the absence of a "real" market and no, an Alice in Wonderland look-alike doesn't hack it) is to work out the likely cash flow on the actual security, not pontificate about spurious rules of thumb. For that you need to drill back down the daisy chain to the actual pool of mortgages or assets and take a rational view of the delinquency rate (and other things). That's theoretically possible, that information ought to be available from the service provider; but that's not all you need.
You also need in the words of International Valuation Standards "sufficient market-derived-data", to be able to understand how the pool of mortgages is likely to perform. That's line-by-line data on the performance of other mortgages or assets.
The problem is that data is not available to the "public" (i.e. you and me, and investors). It is available for stocks, and there are laws that are supposed to make sure people don't cheat, but it's not available for securities.
That data exists. A recent study (.pdf) (coincidentally also of Columbia University), analyzed 700,000 mortgages using line by line data. Not that I agree with the result because they didn't put the mispricing of the housing market in as a variable, but still, it was the first coherent analysis of delinquency I ever saw.
Anyway, I wrote to Dr. Jang Wei asking if he could "lend me" the database so I could test my theory (that mispricing is a valid explanatory variable), and he wrote back very promptly and politely "sorry, the data is confidential" (that's what I thought he would say but I thought I'd ask).
Sure it's confidential, to the bank that owns it. And that has been feeding the market with fodder for melanin tainted toxic assets for five years. If the customers for that junk had had access to that data they would have been able to make rational and informed decisions.
If anyone wants to start to make some progress on sorting out the pile of rotten fish that is the "toxic asset" problem (outside of getting the taxpayer to front up $1 trillion or more so that some Wall Street types can make billions out of PPIP), and if anyone wants to start to have a coherent discussion about a "market" rather than the lunatic concoction that existed before, investors, all investors, are going to have to have access to the data that Dr. Wei has, and similar data from every single bank that ever sold on a mortgage for securitization.
And it should be provided as a matter of Law. Perhaps I am being thick again, but from my perspective not providing that data and selling that junk was a form of racketeering, plain and simple.
With that data anyone who has half an idea about how to do (proper) valuations, could figure out a way to value every single toxic asset in USA. OK, they might be wrong, but that's beside the point, that would give investors the ability to make a rational and informed decision about what to buy and what to sell, and if they got it wrong, well then the only people they could blame is themselves.
Perhaps there is a conspiracy to keep that data out of the hands of the investors? A joke I saw in a letter to the Financial Times a while back went something along the lines of "you don't shear a sheep before you take it to market, lest the buyers can see the imperfections". Well perhaps it's time to let the buyers take a good look at what they are buying for a change?
There is nothing "confidential" about the data, so what if I know that loan number 198-4328-0075 is to a Filipino male, divorced, income $44,000 per year, owns a dog, bought the house in 2006, LTV appraised at 72%, lives in a city, in Michigan; three months late on his installments? That doesn't tell me who he is.
I imagine the rating agencies get more data than you or me, but they are simply gatekeepers of the data.
Their pitch to maintain that monopoly is "trust us we know what we are doing, our ratings are super-reliable."...."Oh and by the way they are just opinions and you should do your own due diligence". Oh yeah, "but trust us"!
Great, so "Give-Me-The-Data" and kindly get out of the kitchen. If Secretary Geithner wants to rely less on the ratings agencies the first step is unlock the monopoly that they have on the data.
Right now the data that is required to do a proper valuation does not exist in the public domain, and for that reason, anyone who is buying or selling toxic assets is simply gambling. That is not investment.
I floated this idea about three months ago and I got a comment saying that the Big Banks like Goldman Sachs (NYSE:GS) and the like needed to value thousands of toxic assets a day, so how could they afford to do proper valuations? My response was "well, they evidently couldn't afford NOT to because otherwise they wouldn't be crawling to the government to bail them out".
With the right data, doing a proper valuation is easy and cheap. That data is sitting on computer databases locked up in the vaults of the shysters who created the mess. Time to cough it up.
One thing is for sure, until they do, no one is going to buy that stuff unless the government pays them to cart it away.
How To Close Down Riyadh International Airport For Two Days:
On the subject of confidentiality and secrets I thought I would close with a story about how my mate Bruce Parker managed to close down Riyadh Airport.
He was looking after a contract digging two meter deep trenches for a 1,000 mm water pipeline through rock. He had a Rock-Saw working, which is basically a D-9 Cat with a huge chainsaw on the back. It used to go through solid rock (the stuff you hit with a hammer and the hammer bounces up and hits you in the face), at about three miles an hour. The ground would shake and the dust cloud would envelope the machine so you couldn't even see it.
He had a team of about twenty people getting all the NOCs, and identifying services in front of the machine. Someone would go down to the telephone people, the water people, electricity, drainage, etc, get the as-built drawings, then he would have someone walk around with a detector. It was a process.
One day he was sitting in his car watching the machine work, and a helicopter flew over, then two trucks full of soldiers came screaming down the road straight past him with sirens blazing, then six police-cars came screaming past from the other direction, then there were five helicopters in the air, and bit by bit the center of activity got closer and closer.
Finally one of the helicopters landed near the machine and some very irate looking commandos with guns started looking in the trench. He walked over, they asked, "who's in charge". That was his second mistake, he said, "I am".
What had happened was when they built the brand new airport they kept all the radar at the old airport and ran a fiber-optic cable to the new one (it was about thirty kilometers away). Since that was essential for the whole operation of the airport they encased it in an one-meter by one-meter continuous slab of concrete. The Rock-Saw had gone through that, and the cable, without even noticing.
In the recriminations that followed, Bruce said, "Well it wasn't on the As Built".
The guy on the other side almost had an apoplectic fit, "Of Course it wasn't on the As Built...IT'S A SECRET!"
That's what caused the credit crunch, too many secrets.
Disclosure: no positions