Jason Cawley's  Instablog

Jason Cawley
Send Message
I am the Director of Architecture at Wolfram Solutions, the consulting arm of the large privately held software company, Wolfram Research. I manage teams of programmers developing custom applications for business and, government, applying advanced analytic methods to practical challenges. I... More
  • Deflation Of The Shadow Banking System 21 comments
    May 30, 2012 8:43 PM

    Anyone can look at the 10 year breaking below 1.7% to set all time lows and notice that we are in a deflation, not the inflation so many predicted. The predictions stemmed from the high visibility of the sectors expanding their sheets - the Fed and the Treasury, to a lesser extent non-financial corporate business with its famously high current profits and large cash positions, which have also less visibly been adding to their total bonds outstanding. So where is the deflation?

    The short answer is domestic financial sectors. In the Fed Z.1 "Flow of Funds" data released quarterly, we can see the total debt of the US financial sector declining by $3.5 trillion since the end of 2008. This has neatly matched the expansion of the Fed's sheet and of Treasury debt outstanding, resulting in the total dollar credit line not moving at all, despite deliberately loose monetary and fiscal policies. Commentators who have paid attention to this contraction of the financial sector have gotten the basic call on the bond market direction, subdued prices and weak economic growth, correct, while anyone who missed it expected high inflation from the policy response lever-positions, not seeing the condition those levers are reacting to.

    So this is the deleveraging everyone talks about but few seem to understand in detail. OK, broad credit not moving because the financial system has not healed from the damage inflicted by the 2008 crisis (and preceding bubble), check. But where specifically has the contraction occurred? What institutions and line items are we talking about?

    The first thing to notice is that it is not the major money-center banks, nationally chartered or bank holding companies (on balance sheet). Their (direct) debt out has expanded by a little over $300 billion, from $1.4 trillion to $1.7 trillion. (All figures contrast the end of 2008 level to the last quarter of 2011, the last available full quarter of Z.1 data). The smaller savings and credit unions have cut their debt out by $270 billion over the same stretch, so this is really a size consolidation more than any growth of the functional sector.

    The next thing to notice is that it isn't the agencies, primarily, though they are a part of the story. They have taken pool assets back onto their sheets over this period, so the meaningful comparison is the sum of the line items for GSEs themselves and for GSE mortgage pools. Those combined stood at $8 trillion at the end of 2008 and a little over $7.5 trillion now. So this is part of the contraction, $500 billion or about 1/7th of the total, and reflects low mortgage origination activity, loans running off into cash, and defaults.

    The big items are, instead, ABS issuers, which is a Fed category that includes all the private mortgage pools, plus the line items for finance companies and funding corporations. Collectively these are the "shadow banking system" - the whole network of off balance sheet special purpose vehicles, loan originators and securitizers, and the funds holding the paper those institutions create. Specifically, the debts out of the ABS issuers category fall by $2.07 trillion, and of the other two combined, by another $1 trillion.

    What are the internals of those? For that we need to turn to separate releases specifically about the ABS issuers, release L.126 in Fed-speak. On their liability side, we find they retired $500 billion of their commercial paper and $1.6 trillion of their bonds. On their asset side, we see the disappearance of $900 billion in home mortgage assets in the SIV pools, another $100 billion in commercial mortgages, and $530 billion in consumer credit receivables, both CC and other consumer loans. There are smaller reductions in their other line items, like trade receivables, but those are the bulk of the changes. The finance and funding corporations similarly show large run-offs in their own commercial paper combined with asset reductions from consumer loans, car loans, small business receivables credit, and the like.

    These reductions amount to cutting the shadow banking system *in half*, since the crisis. The changes are larger than all additions to the Fed's sheet by a factor of 2, and are larger than the modest reductions in debts of the household sector, by a factor of 5 or more. The ABS market basically is no more, and its existing assets are in run-off. While the rate of decline from the evaporation of these assets and liabilities, both, has fallen since its peak rates of 2009, it remains substantial, on the order of $500 billion per year, about half of it in the ABS issuers line item alone.

    So there is the deflation. This stuff isn't coming back, but as long as it is shrinking, total credit out is not moving higher. Basically the end owners of all of these SIVs are using all available free cash flow from their run off to retire the debts that financed them in the first place. It is hard to see where serious net credit expansion can occur before this stuff has finished being cleaned up, though the headwind from its runoff is diminishing with the remaining size of the sector.

    I hope this is interesting.

    Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

Back To Jason Cawley's Instablog HomePage »

Instablogs are blogs which are instantly set up and networked within the Seeking Alpha community. Instablog posts are not selected, edited or screened by Seeking Alpha editors, in contrast to contributors' articles.

Comments (21)
Track new comments
  • No Free Cake
    , contributor
    Comments (1300) | Send Message
    Thanks for your continued efforts Jason.
    30 May 2012, 09:28 PM Reply Like
  • Chambord
    , contributor
    Comments (78) | Send Message
    Excellent as always. http://1.usa.gov/KK7Ohg
    There's a whole lot of distrust in these numbers. And a bunch of people who can pay down or pay cash.
    31 May 2012, 12:00 AM Reply Like
  • Jason Cawley
    , contributor
    Comments (5473) | Send Message
    Author’s reply » Thanks for the link. I don't distrust the numbers, but it is necessary to look at them alongside the major banks. Because over this period some of the SIVs have been moving back onto the balance sheets of sponsoring institutions. As I mentioned in the piece itself, though, liabilities of the major banks have moved up only $300 billion over this stretch (and S&Ls shrank by almost that much), so migration from one accounting line-item to another does not account for the shrinkage. To the extent that stuff has been leaving these forms, it has been to be liquidated, here or after migrating to other line items.


    Most of the assets in these vehicles have relatively short weighted lives. Even the mortgages often have run-off rates into cash around 20%, particularly when rates are moving lower, encouraging refinancing of existing loans. The consumer stuff has evaporated by 5/6ths, which over 3 years would about match run off without any new issuance, in those shorter term loan classes.


    So my "read" on the mechanisms behind the decline starts with a near-complete absence of renewals using this form, some migrating back onto major bank sheets and seeing the liability side paid down there out of overall bank cash flow, some lost to default - but most repaying more or less as contracted, with all returned principal allocated to paying down the bonds and commercial paper issued by these entities to finance their asset positions.


    Both the assets and the liabilities are no more because they have been repaid. Net liquidation of debt through repayment is the main mechanism behind deflationary periods. In the previous bubble period, net new issuance to the tune of $1.5 trillion a year was making all other actors and sectors combined feel flush; now the reverse to the tune of $1 trillion a year initially, half that now, is making all other actors and sectors combined face tightness of credit and diminshed demand.
    31 May 2012, 11:39 AM Reply Like
  • Chambord
    , contributor
    Comments (78) | Send Message
    I meant distrust in another way. No one trusting the CP market anymore -- there must be some good names out there like V or MC issue CP? Corporate borrowers not trusting the CP market so they borrow long. Banking not trusting individual borrowers, small corporates, reits, etc.
    31 May 2012, 12:10 PM Reply Like
  • Jason Cawley
    , contributor
    Comments (5473) | Send Message
    Author’s reply » Ah, yes, quite, I think the "discredit" flow went from end purchaser to ABS and SIV issuers and only through them to ultimate borrowers. Yes there was such discredit, distrust of certain categories of issuers that restricted new issues and forced run offs. I think it went, SIV trying to issue CP finds it cannot do so without an unacceptable concession in yield, so it needs to let its own loans run into cash as its CP matures, to repay it on time.


    The ABS issuers themselves could not place their liability-side paper. Credit flowing to small business by sale of its receivables, or the ability to resell CC receivables for immediate cash, therefore declines. This makes all the finance and funding corporations less liquid - they can't turn new loans into immediate cash as quickly as before, as their whole business model was set up to do, in fact. Instead they find themselves holding more of the loans they originated than they want to.


    In the short run they probably have to run up their letter of credit lines with their banks to compensate; in the not-much-longer run they slow or halt new loan originations. Some of the loans migrate to the money center banks as collateral; they don't particularly want it and leave it in run-off.


    For the mortgage vehicles, they can't place new bonds because the potential purchasers are unsure if this or that SIV is high risk and just avoid the entire sector. New issues don't get funded. Existing ones need to repay the current portion of their long term debt out of cash flow. Perhaps some even see their debt fall to discounts in illiquid secondary markets and find the best use of cash flow is to buy it in early. Others hit funding snags when they can't rollover maturing CP, and cough collateral to sponsoring banks, or get brought back on-sheet.


    It wasn't that end purchasers initially decided that consumer loans were losers and they didn't want to lend to anyone making consumer loans. They didn't trust that the obscure SIV counterparties were not Lehman style disasters waiting to happen, enough people shunned such paper, the vehicles couldn't roll their liability side stuff as a result. Their asset side stuff has to be realized ASAP. That cuts off credit flowing to their end borrowers, whose standing falls as an indirect result. (The small business that can't sell its receivables becomes more likely to fail to meet its payroll and go bust, etc).
    31 May 2012, 01:30 PM Reply Like
  • Westcoaster
    , contributor
    Comments (750) | Send Message
    Great post.


    Is this mainly reflected in the intermediaries being gone? Lehman, Bear Stearns, and AIG?


    What an amazing deleveraging we have been through. Also isn't 2 or so trillion of this now sitting on the books of the Fed?


    Thanks in advance....
    1 Jun 2012, 01:09 PM Reply Like
  • Jason Cawley
    , contributor
    Comments (5473) | Send Message
    Author’s reply » There are intermediaries that are gone, but no not Lehman Bear or AIG particularly. Instead a lot of little SIVs created by every major bank and Wall Street firm. Certainly in the case of non agency mortgage backed securities, the failures of Lehman and AIG destroyed the market for that stuff - AIG wasn't around to write CDS on such paper which made it "uninsurable", and Lehman having been driven under by having too much of it on their sheet when its prices cratered, dissuaded others from holding it and from having a big pipeline originating it. The nationalization of the agencies also helped shut down the private MBS market. But that is specific to the mortgage side of things. 40-45% of the deflation is in mortgages, about two thirds of that in private MBS run off and the other third in agency losses and run off. But the rest of the ABS market went with it.


    As for the question, is this paper now on the Fed's books, the answer is "no". The Fed did create special loan facilities in late 2008 to finance specifically asset backed commercial paper when it is most stressed, and also a larger facility to support money market funds, also for a while. But these were short term loans on money market collateral, basically, and liquidated themselves by mid 2009. Term auction credit to the major money center banks took over from all that, as the banks became willing to lend on the CP still out in early 2009. Recall that at the time, major regional banks (2nd tier, KeyCorp as an example) saw their short term notes (2-3 years) hit 10, 12, and even 14% in the secondary market, so there was extreme distrust of all financial issuers (unjustified, as it turns out). The Fed was basically acting as a traditional lender of last resort on collateral that was basically sound but that a panicked market distrusted.


    All of that unwound during calendar 2009. The banks repaid all of it as their own credit was restored, which can be timed to the turn in the corporate bond market around March to April 2009. By October of that year, the Fed was plowing the proceeds of these repayments into agency MBS, to keep its sheet the same size.


    The only non-agency MBS on the Fed's sheet in any of that were portions of the assets of the three Maiden Lanes, the first for the Bear failure and the others related to AIG and Citi. Those amounted to about $80 billion initially on its sheet. Understand these were pools of assets purchased at steep discounts to face - 50% typically - that the Fed agreed to lend against at those lower prices to support those clean-ups. With a bank counterparty standing ahead of it for losses on the pools. These have been running off and getting repaid and now only amount to $20 billion on the sheet, and are about $5 billion in the black. That mostly just reflects the low prices paid for them, the partial recovery of the market since, and the low crisis prices proving lower than actual realizations on the underlying loans.


    The Fed did buy $1.1 trillion in agency MBS by mid 2010, first from the repayment proceeds on all the short term credits from the crisis period, and later from "QE 2". Then it let those go into run-off, pretty much, and about $300 billion of it came back as cash over a little under a year and a half (through the end of 2011). It currently has $840 billion of agency MBS, having added to the position in line with the whole sheet size since the end of last year.


    But these agency MBS are not the evaporating assets of the private ABS issuers. They are mortgage pools from Fannie, Freddie, and Ginnae Mae (HUD), and since those are all government owned at this point, they all have treasury guarantees. Not risky. The total of all the agency paper out has fallen by $500 billion over this stretch, so it isn't a run-up financed by Fed purchases. Instead it is a shift of existing treasury-guaranteed mortgage bonds from the hands of banks (mostly) to the Fed, in return for which those banks have big extra reserves on deposit with their Fed branch. In effect, the banks sold a bunch of their bond portfolio to the Fed to run up their checking account balance, to get more liquid and lower their risk level.


    The overall expansion of the Fed's sheet since the crisis has been just under $2 trillion ($44 billion shy of that). Most of that is purchase of treasuries (up $1.175 trillion), the balance the MBS described above. The treasury has added a little over $4 trillion to debt out, a quarter of it going to the Fed. The private financial sector shrank $3.5 trillion in the same period, and the household sector reduced its debts by $500 billion, basically counteracting these policy expansions.


    If you look at the Fed Z.1 debt levels table and compare total non financial sectors (which includes corporate and non corporate business, state and local governments, as well as the above) plus financial sector, at the end of 2008 the total dollar debt out (US issuers) was $51.588 trillion. As of the latest release, the end of 2011 4Q, the same total is $51.895 trillion. Total dollar debt out has thus increased only $307 billion, or 0.6%, over 3 years - basically flat. Policy both monetary and fiscal has been very loose, but has only offset the contraction of the private financial sector being discussed here, for essentially no net change in debt out.


    I hope this helps.
    1 Jun 2012, 01:45 PM Reply Like
  • No Free Cake
    , contributor
    Comments (1300) | Send Message
    Jason, forgive my ignorance but can you help clarify something.


    You describe the Fed buying the agency MBS from banks. But, isn't that why people say the Fed now owns "toxic" assets that banks used to hold?


    I think the term toxic is just used to describe poorly performing, constructed, and/or underwritten loans very broadly. And, some of these were held by the agencies. Which, before the smash, were GSEs but without *explicit* Treasury backing.


    In this sense, haven't toxic loans been "transferred" to the Fed?
    1 Jun 2012, 08:31 PM Reply Like
  • Jason Cawley
    , contributor
    Comments (5473) | Send Message
    Author’s reply » The term "toxic assets" refers to assets suffering declines in value expected to prove permanent, either through credit defaults or market reprincings that aren't temporary - which could include a permanent reduction in the credit standing of a particular class of loans that causes it to trade at wider spreads, for example. The point behind the terms is that all insitutions expect to pay for their capital on the liability side (borrowing to fund their asset positions), but expect to earn on their asset side, more than enough to cover their cost of capital. When instead the asset side posts net losses, we speak of "toxic" assets.


    Notice that routine default losses lower than interest rate spreads would not qualify - e.g. it is entirely normal for credit card loans to have default rates of 5-6% of principal in good times, and up to twice that in bad ones, but that is why they charge 13-21% on open balances. Credit losses on that scale are covered by the interest charged, the asset still earns something net, and that isn't "toxic".


    During the 2008 crisis, specifically subprime mortgage bonds were an example of toxic assets. Subordindated tranches of even prime mortgage loans likewise, as first exposed to default risks. Some banks (notably Citi) had believed that the senior tranches of subprimes were sounder and were "caught out" when this proved a delusion - they had expected that 20-30% cushions above them from subordinated tranches would protect them from default risk. But in the event, even the senior parts of such pools fell to 30 cents on the dollar at the market bottom, before recovering to 45-50 about a year later.


    But agency MBS were never in this category. Their prices have never been impaired, in any of it. At this moment, for example, a FNMA 4.5% initial loan rate MBS trades at 107.3 - the 4s trade at 106.5 and 5s trade at 108.3. Holders of agency MBS have received on-time payment of interest and principal exactly as contracted. Because mortgage rates have moved lower, loan pools at previous higher rates have gone to premiums, and their average life has fallen significantly due to refinancing activity. The current weighted life of a FNMA 4.5 pool is only 2.5 years, for example. The cash flow a holder of such a security receives is thus far above the initial purchase yield, because the principal is also being returned quite rapidly.


    Why have agency MBS shown no losses to their holders, when default losses on average mortgage loans have ran as high as 2.85% per year at the recession peak, and are still running 1.4% or so? Because the writing agency insured the loans - that is what the agencies do. The agencies themselves can run losses if credit losses falling to them (after any PMI ahead of them, net of recoveries, etc) exceed what they charge in insurance premiums, and that has happened. They have fufilled some of their insurance obligations by repurchasing individual mortgage loans from their loan pools, taking those on their own sheet - which is an indirect way of meeting their contractual obligation to guarantee payment of the principal.


    But none of that has affected the end owners of agency MBS. Whether the Fed since its purchases, or banks that owned the stuff before, or mutual funds that invest in it, or retail direct owners of agency MBS - they have all been paid on time on all of it, and in addition have seen the market price of their loans rise not fall. These securities are also liquid, they can be sold at any time at those premium quotes, in size, without any problem.


    They do not meet the characteristics of a toxic asset, therefore. Nobody saw their own organization's internal finances destroyed because they were paying interest on a liability side to carry agency MBS, then the asset side MBS gave them losses instead of income, impairing payment of their liability side.


    The banks sold a portion of their bond portfolio that was trading at full prices because it was completely unimpaired, to the Fed, for cash reserves, becoming more liquid in the process. They gave up income in doing so, and they missed a capital gain on the appreciation in price of the agency MBS as mortgage rates fell. By holding $1 trillion in excess Fed reserves for about 2 years instead of the same amount of agency MBS, the private banking system lost out on about $85 billion in earned interest, and another $70 billion or so in a missed capital gain. Those gains were earned by the Fed, instead, with the earned interest portion returned to the US Treasury in the Fed's annual payments, and the capital gain portion still unrealized and part of the Fed's operating capital.


    The only thing the banks got out of it was they became more liquid and lowered their interest rate risk. Since rates moved lower not higher, that lower interest rate risk did not prove to be a benefit - and the Fed forecast that movement correctly, rather easily, being partly responsible for bringing it about. Being more liquid was meant to increase confidence in the banks, helping them raise private capital on better terms, both with stock sales and in terms of the market rates on their bonds. It undoubtly helped somewhat in all that, at the margin, but the major banks are still trading at discounts to book value and paying more than other corporations to borrow.


    Presenting any of the above as the Fed absorbing losses to spare the private banks is the opposite of the truth about what happened. It absorbed large gains that the banks would otherwise have earned, by taking on a normal and profitable banking role that the banks were performing before it moved in. The macro effects of the Fed being easier in all of this doubtless help the banks, as they help the whole economy and the banks benefit when the economy does. But the direct effects of the Fed holding agency MBS instead of the banks' holding them, earns for the US treasury at the expense of private banks. It takes over some of their ordinary profitable activities and keeps the profits of those activities.


    Fair question.
    2 Jun 2012, 02:09 PM Reply Like
  • Westcoaster
    , contributor
    Comments (750) | Send Message
    Thanks JasonC, you won't read that explanation in the paper.


    So you would say Andy Beal didn't need to warn the Fed or the Fed listened to Andy and they didn't put garbage in the Fed? That is just great news.


    Thank you....


    2 Jun 2012, 03:24 PM Reply Like
  • Jason Cawley
    , contributor
    Comments (5473) | Send Message
    Author’s reply » I'd say the word "mortgage" apparently confuses people and too many can't tell one thing from another if a single English word is shared by the longer names of the two things.


    The concern in March 2008 refers to the Bear bailout and specifically to the formation of the first Maiden Lane vehicle, and applies to the later 2 Maiden Lanes as well. But it emphatically does not apply to *agency* MBS. It is a bit like thinking treasury bonds are risky because junk bonds proved risky in the 1990 recession and both share the word "bond".


    The 3 Maiden Lanes did include assets that had been "toxic" to the holding bank. The first dealt with the bankruptcy of Bear Sterns, the later 2 with asset pools from AIG and Citigroup. The structure was similar in all three cases - a pool of toxic assets was sold to the ML entity at a steep discount to face value; a bank counterparty (JP Morgan Chase in the case of Bear or ML 1, as the follow-on purchaser) took a subordinate position ahead of the Fed for any losses on the pool, and the Fed lent against the assets on that basis. Besides its share of the interest it was to get either all of any excess of realized value over the initial (discounted) price, or share it 50-50, varying from one of the MLs to the next. The loan pools were left in run-off with the partner bank acting as loan servicer.


    Fundamentally, these were clean up operations to help the saved institution do other deals by removing uncertainty produced by those assets being on their sheets. In the case of Bear it convinced or enabled JPM Chase to buy what remained of Bear; in the case of Citi and AIG, it enabled the treasury to buy their preferreds as part of TARP with these problem assets removed beforehand.


    These are the only operations in the crisis that fit the description of toxic assets moved to the Fed's sheet - but do not refer to the later trillion dollar operation in agency MBS. The timing is a year earlier, and the scale is an order of magnitude smaller - $80 billion at the peak of the ML balances. Which has reduced by loan run-off and repayments to $20 billion now, which are $4-5 billion in the black.


    Commentators at the time objected to the MLs, some on principle and precedent, and others more prudentially worried that the assets put in the pools wouldn't actually prove to be worth enough, so the Fed would lose money on them. They were worried that the Fed would get in the habit of monetizing failed assets and sticking the Fed and indirectly the treasury with the losses.


    Those fears may have been reasonable at the time, and the points about principle and precedent may be sound even now. But in the event the MLs have mostly liquidated themselves naturally and they haven't cost the Fed or treasury anything. This is mostly a function of the steep discounts put on the assets when they went into the pools, typically 50% of face. Those were market prices at the time, and reflected crisis fears. The actual experience of the loans in terms of repayment and credit losses etc have certainly left them worth well less than their original face value, and in that sense they were always toxic assets to their originating bank. But they took that 50% hit when the MLs were formed - the Fed certainly didn't pay full face value for them to spare its counterparties or anything like it.


    Could they have lost even more than those discounts and left the Fed with losses? Potentially, but it didn't actually happen. They would have had to prove worth less than the original discounts, and still have gone through the capital the counterparty banks contributed to the MLs (they were the subordinate, the Fed the senior position, in those), etc. It was conservatively financed but in dodgy underlying paper; the conservative financing put the actual losses from face on the counterparty banks. They benefited from downside protection, and from immediate liqudity - they couldn't have easily sold the assets in the middle of the crisis.


    But when commentators try to pretend that the whole agency MBS position, built up in from the spring of 2009 to mid 2010 and reaching $1.125 trillion, and still $840 billion, are "toxic assets", that's nonsense. They are either themselves confused or are exploiting audience confusion about the distinction between the small Maiden Lane operations of 2008 - $80 billion then, $20 billion now, $4-5 billion in the black - and that larger mortgage security investment.


    Again a fair question...
    2 Jun 2012, 04:12 PM Reply Like
  • No Free Cake
    , contributor
    Comments (1300) | Send Message
    Thanks Jason, your explanation will help me separate the wheat and chaff now.


    I agree that many people use a more colloquial definition for "toxic" than the stricter banking term you describe. I certainly was.


    I hope you continue to be active on this board. Without your clarity I could be easily led astray by (supposedly) learned commentators.
    4 Jun 2012, 10:47 AM Reply Like
  • Westcoaster
    , contributor
    Comments (750) | Send Message
    Thanks Jason...


    I guess I am confusing my Maidens Lanes with my Quantitative Easings.


    I'm glad they didn't put that much garbage in the Fed after all.
    2 Jun 2012, 08:51 PM Reply Like
  • Alex_G
    , contributor
    Comments (1116) | Send Message
    Jason, just read your instablog and all comments. Might be the finest blog and follow up I've read on SA. Thank you.


    29 Jun 2012, 11:45 AM Reply Like
  • crozz
    , contributor
    Comments (297) | Send Message
    I agree. Exceptional.
    15 Sep 2012, 08:47 PM Reply Like
  • dnorm1234
    , contributor
    Comments (1111) | Send Message
    Agreed. Internet commentary doesn't get much better than this guy.
    27 Sep 2012, 05:50 PM Reply Like
  • Jason Cawley
    , contributor
    Comments (5473) | Send Message
    Author’s reply » Thanks for the kind comments. I want to find time to write more of these...
    28 Sep 2012, 11:43 AM Reply Like
  • Westcoaster
    , contributor
    Comments (750) | Send Message
    John Mason's article sort of contradicts your instablog.




    Any comments on FSB's claim that shadow banks have grown to 67 Trillion?


    Thanks in advance.
    19 Nov 2012, 02:03 PM Reply Like
  • Jason Cawley
    , contributor
    Comments (5473) | Send Message
    Author’s reply » It is a fair question, let's look at it.


    To start with I didn't believe the headline. I saw the story in the financial press too, but figured they were obscuring the decline in the US with some other foreign developments, I wasn't sure what. So let's go to the actual report instead of an internet AP blurb.


    You can find the whole thing here -




    The highlights -


    "Compared to last year's estimate, expanding the coverage of the monitoring exercise has increased the global estimate for the size of the shadow banking system by some $5 to $6 trillion."


    Notice, they do not claim that shadow banks are expanding. They claim that a wider coverage of shadow banking puts its total size larger than a narrower previous coverage area. To get what this means, they simply didn't have any data on Switzerland, Hong Kong, or Brazil in the previous totals. They are just including more country's data in the totals, not recording organic growth of those totals.


    "The shadow banking system's share of total financial intermediation has decreased since the onset of the crisis and has remained at around 25% in 2009-2011, after having peaked at 27% in 2007. In broad terms, the aggregate size of the shadow banking system is around half the size of banking system assets."


    Notice, they say the share of the financial sector represented by their broader definition has fallen, not increased, since the crisis, and that it remains about half the size of the ordinary or regulated banking system. They just claim it hasn't continued shrinking since 2008.


    Their new definition is all financial holders or issuers of credit products other than commercial banks, central banks and state enterprises, insurance companies, and pension funds.


    They agree that the specifically US shadow banking system (even so defined) has fallen, since they state it is now 35% of the global total, down from 44% in 2005. That is a 20% fall. The US had the greatest share of its financial system in those categories before the crisis, and has experienced outright contraction of that portion.


    The total figures and US portion they give are $62 trillion worldwide in 2007, $59 trillion in 2008, and $67 trillion in 2011 ($23 trillion US).


    Their figure 2-3 shows a graph of shadow banking assets to GDP, and it declines 15% from the 2007 level - a sharp downtick in 2008, recovery in 2009, and then a slide in 2010 and 2011 back below the 2008 low.


    They arrive at these figures by including ordinary mutual funds in "shadow banking", along with ETFs and similar products. Including equity funds, not restricted to credit specifically. Quoting from the report -


    "The largest sub-sector, representing $19 trillion and 35% of assets of NBFIs in 2011, is that of "other investment funds", i.e. funds other than MMFs. This sub-sector is very diverse in its risk characteristics and includes equity funds, bond funds of varying degrees of credit risk, mixed funds as well as Exchange-Traded Funds (ETFs)..." A later chart shows that 53% of that balance is equity funds and another 14% is in balanced funds - so only about 1/3rd of that total is in credit products.


    Mutual funds, especially those holding stock, are not what you and I would call shadow banks. (Also notice they'd rise in "amount" just from stock markets doing well). Only about 1/3rd of their total are line items we'd call shadow banking - structured financial vehicles, financial holding companies, finance companies, US funding corporations, and Dutch covered bond arrangements. Another sixth, about, is split between money market funds and broker-dealers, which might or might not rate as shadow banking (basically all of them are in the US and Japan, which have separate investment banks that underwrite securities, vs universal banking in most of the rest of the world), and the last sixth (about) is just "other", not falling into any of the above. They acknowledge that some of that may just be uncovered jurisdictions (Brazil Switzerland Hong Kong etc fell there before they included them this time around), some may be hedge funds otherwise undercounted, and the like.


    Conservatively we should back out a third of their total, maybe half, maybe even two third of it. If we back out half of it we'd minimize the error bar, but it would still be large. That puts the shadow banking system at maybe a quarter the size of the official banking system, maybe a third at most.


    Then they report that structured financials contracted 12% in the last year alone, money funds contracted 5%, while the other categories were basically flat. Quoting from the full report - "The trends in 2011 appear to be a continuation of the stagnating or declining patterns observed since the crisis in an environment of elevated risk aversion and muted financial innovation." Note the glaring difference between this assessment, on page 18 of the full report at the end of a paragraph, and the headline the financial press reported.


    Where is there actual expansion? Emerging markets, especially inflation prone ones - "certain emerging markets (Argentina, Brazil, China, Indonesia, Korea) have experienced strong growth in several non-bank sub-sectors since 2007" - and the accompanying table would add Russia and India to that list. Canada has seen some growth in mutual funds and ETFs, about all there is in the plus column in any of the developed world.


    The shadow banking system is not expanding. Not in the US - declining sharply - not in the rest of the world. In a few developing countries there is rapid growth from small initial levels. Those monitoring it are putting more and more things into the category, both in width (more countries) and depth (now a stock mutual fund is "shadow banking" for some reason). Then the headline writers pretend bubble blowing off the record is still happening, when it flat isn't.


    Why do journalists so deliberately get these things wrong? For the same reason no news show will ever run a story on monetary policy without showing a printing press making dollars. It is populist yellow journalism, and slander of finance is the grand narrative into which everything under the sun must be crammed. Never believe a journalists headline - instead track things back to the figures themselves and analyze what is actually happening. You will find it is practically never reported straight.
    19 Nov 2012, 06:07 PM Reply Like
  • Westcoaster
    , contributor
    Comments (750) | Send Message
    Thanks Jason, I think I was mixing my Shadow Debt with Shadow Total Assets. Your article above discussed the deleveraging and as this FSB report is talking about total assets.


    Sorry bout that....


    Any commentary from you on the point of this report?


    Are you afraid of your shadow....bank?
    21 Nov 2012, 12:14 AM Reply Like
  • Jason Cawley
    , contributor
    Comments (5473) | Send Message
    Author’s reply » LOL - thanks. No I am not particularly afraid of shadow banks.


    I do think the model of having all assets in exchange traded form and then thinking one must thereby be liquid because repos can always carry them / fund them short, was unsound to start with and the 2008 crisis revealed that fact. Banking relationships cannot rely on market prices always being sane, markets always being open and clearing with stable prices, etc. Now, that can still be handled by other methods besides being a commercial bank - low enough leverage, enough of a position funded with equity not debt, a portion of the assets short-term, safe and liquid themselves - etc. But the aggressiveness with which they were being run in the "oughts" was unsound, a typical "bubble" practice like pyramiding in 1929 or conglomerates with layered leverage in the 1960s, or all junk-debt financed LBOs in the 80s. I simply don't expect those practices to come back.


    The issue the collapse of the speculative shadow banks raises for the whole system is just the depression of broad credit that goes with getting rid of them, and lowering overall financial leverage to make the financial system safer. Something has to replace the missing broad credit they were providing, and until other forms of credit grow enough to fill their previous role (more soundly, we hope), macro performance will remain weak. Simply because broad money growth is effectively non-existent, until we get through that de-leveraging.


    Anything that makes it pay to add financial capital - in equity form, not more debt and especially not volatile short stuff - to the broad financial system, will speed up that process.
    21 Nov 2012, 12:36 PM Reply Like
Full index of posts »
Latest Followers

Latest Comments

Instablogs are Seeking Alpha's free blogging platform customized for finance, with instant set up and exposure to millions of readers interested in the financial markets. Publish your own instablog in minutes.