Further memorandum from Odey Asset Management
We judge there is a significant opportunity
for government to restart the credit flows through the financial
and real economy of the UK, to make that happen fast, to ensure
fair value for tax payers, and to do this with an exit strategy
We recommend a two-stage package of UK government
a national bank for troubled debts
or government guarantee scheme; and
active intervention to kick-start
corporate debt markets.
... other interventions will be necessary as
the crisis progresses and should be coordinated with this package.
Without this scheme as a foundation, existing and future measures
look likely to fail.
We also offer a comment and proposals on HM
Treasury's latest guidance on the proposed asset protection scheme
to deal with bad debts (page 3).
a recent IMF working paper* showed
that 60% of all global banking crises in the last 30 years (42
crises in total) were ended by a bad bank or troubled debt guarantee
structure. This structure is missing from the policy response
in the UK today;
the buyer of last resort approach
for company debt is analogous to the US TALF intervention scheme.
The TALF provides evidence that this approach delivers significant
positive effects for the wider economy and population;
international experience provides
elements of successful road-maps for implementing the proposals;
these measures would not hand UK
banks a "free lunch". It will ensure that solvent UK
banks remain in private sector control;
the two building block measures will
work best when accompanied by other sensible policy measures including
credit guarantee schemes, and the much needed flexing or suspension
of those Basle 2 rules which have a pro-cyclical impact; and
there is no reason for government
to make a loss over the life of the project.
Stage One: create a National Bank for Troubled
Debts or government guarantee scheme
1. A bad bank or troubled debt guarantee
scheme is an essential part of a successful healing process to
this banking crisis.
2. The fundamental problem is too much debt.
This is exacerbated by the fact that all investors and lenders
know that the UK banks have bad debts inside them still to come
out, but they do not know how bad those debts are. As long as
opaque, impaired assets remain on bank balance sheets the negative
feedback loop of deleveraging and losses will continue. This means
that markets will remain dysfunctional, credit availability scarce,
and the banks will be focused on reducing their exposure through
3. De-leveraging is ultimately quite simple,
either you attract more equity or you shed assets. Without balance
sheet repair, it is currently near impossible for many UK banks
to raise equity capital. Therefore they have to de-leverage, reducing
lending to companies and households.
4. Banks in the EU are now valued at below
their audited measures (tangible book value or net assets) which
makes it very difficult to raise new capital. Barring full-scale
nationalisation of swathes of the banking sector at huge cost
to the taxpayer they will therefore de-leverage. This is the negative
death spiral in which we currently find ourselves. So if banks
cannot raise equity we must help the sector shed troubled assets.
The most transparent, efficient and scalable way to do this is
a national bank for troubled debt (a bad bank structure).
5. The banking crises in Sweden in the 1920s
and the early 90s make a relevant case study. To quote the Governor
of the Swedish Central Bank in 1997** (the Riksbank) the problems
of the early 90s "seem to have been more extensive than those
which arose in Sweden in the early 1920s". Yet in the 1920s
the fall in GDP totalled 20% versus around a decline of "only"
6% in the 1990s. Why the difference? In large part because "the
two periods differ substantially in the management of the crisis".
The early 90s Swedish model is one of prompt, transparent handling
of the banking sector problems following the trusted route of:
6. The mechanism for establishing the price
of an asset as it enters the bad bank, coupled with the extent
to which the banks could participate in any eventual upside are
key points in the process. But the models (details below) show
solutions are possible, whether at strategic level the means is
(a) a separate bad bank or (b) guarantees issued by government,
in exchange for premiums paid by the banks, but with the now underwritten
assets staying on banks' balance sheets. The troubled assets are
currently so realistically marked to market that this scheme could
operate on the calculation that government can avoid a loss over
the life of the scheme. If any UK banks chose not to participate,
negotiating for higher prices from government, then that would
be left up to them. All UK banks would be treated equally.
HM TREASURY INITIAL
7. We welcome the fact that the government
is considering an asset protection scheme, however the lack of
detail in the initial 4 page release from 19 January is disappointing.
8. We assess that markets, investors, and
savers will judge it essential to have the detail fleshed out
and the policy implemented, at the very latest, alongside the
full-year results of the banks in mid/end February 2009.
9. Waiting until co-ordinated agreement
has been achieved amongst the G20 is likely to have further very
negative consequences for the UK banks.
10. It is not only the share prices which
have fallen, but the value of subordinated debt instruments issued
by the banks have in many cases reached badly distressed levels.
This creates risk for pension holders as the insurance sector
holds substantial amounts of this type of bank debt. For example,
Legal & General and Prudential collectively own £4 billion
of subordinated bank debt.
11. Our analysis shows that the reduction
in risk weighted assets (RWAs) which the banks could gain from
a guarantee scheme would be highly beneficial. But since the public
policy objective is to make bank balance sheets sufficiently strong
for them to substantially increase lending, additional measures
will be necessary. We recommend that the Committee consider the
(i) the first loss tranche of 10% to be borne
by the banks, according to the Treasury plan, will need to be
booked against FY09 profits and staggered throughout the year,
if it is not to further destabilise the banks;
(ii) the further residual exposure of c10% of
the credit losses which exceed the first loss amount must at least
be staggered through 2010-11;
(iii) the government must recognise that receiving
a fee for the guarantee reduces the ability of the banks to raise
lending in the short-term. The two policy objectives are directly
opposed. We think it better that the policy measures achieve their
prime objective, and therefore counsel against sending more mixed
messages to the banks, and against giving with one hand while
taking with the other; and
(iv) for the guarantee scheme to be effective
it has to fully address market fears over further heavily dilutive
recapitalisations or nationalisations at a future date. The means
to do this, are to add the above measures to the package, and
to demonstrate that banks will be able to significantly improve
profit margins so that capital can start to rebuild "organically"
via retained earnings. We suggest the optimum way for banks to
achieve this is to purchase corporate debt, which will help fund
companies and in the current climate offers banks significant
profitable opportunities. In short, for some of the UK banks to
remain going concerns it is necessary that their bad debts are
dealt with; that the costs of that are distributed over a period
of years because the banks' capital is too fragile to pay for
it upfront; and that it is made possible for investors to see
how banks are going to earn their way back to health in the coming
12. What is imperative is that policy-makers
act at speed.
Stage Two: intervention to kick-start corporate
13. This part of the package has elements
similar to the steps taken under the TALF scheme in the US. The
Term Asset-backed securities Loan Facility (TALF) scheme is bringing
down the cost of mortgages to American home-buyers. It is a $200
billion facility that helps market participants meet the credit
needs of households and small businesses by supporting the issuance
of asset-backed securities (ABS) collateralized by student loans,
auto loans, credit card loans, and loans guaranteed by the Small
Business Administration (SBA). It is all about changing practical
things in the real economy for the benefit of the public at large.
14. In our view, a European model would
work best via purchases of corporate debt by the UK government,
either through a new agency founded for the purpose or, and this
would be most effective, through the UK banks. It could also function
15. In outline, the government provides
financing, for a fee, to the UK banks (or to a new agency) which
can be used under this scheme only to purchase corporate debt
in the secondary market. The effect of this is to reduce credit
spreads, thus making it possible for companies to borrow again
in the primary market at sensible prices. The benefit for the
banks is that they will be able to make a profit margin on the
secondary market debt which is higher than their existing business.
This may require an element of compulsion being exerted on the
banks. However, the margin uplift they receive would be substantial
enough, we estimate, to more than cover the most extreme impairment
scenarios on the banks' real economy loan books as the recession
bites. But without the bank for troubled debts, many UK banks
will be in no position to be purchasing more debt in secondary
markets or to make fresh loans.
16. So under our scheme, companies get cheaper
credit, banks get to become going concerns once again, the virtuous
circle benefits the wider economy, and public policy objectives
can be achieved.
17. The two measures in the package are
interdependent. Without such a transfer of bad assets banks may
be encouraged to defer the reporting of losses for as long as
is legally possible. The Japanese banking system of the 1990s
and the US post Savings and Loans crisis point up how such lack
of transparency compounds and extends the problem.
18. Policy-making, thus far, has been far
more proactive than the 1930s, yet credit spreads are comparable
to those in the 1930s (see Fig 17 below). Intervention in corporate
credit markets is necessary to resolve the crisis.
Facts and models for a troubled assets bankwhat
we have so far
(i) US TARPtroubled asset relief program,
size $700 billion
The TARP scheme embraced the concept of a `bad
bank' since its initial purpose was to purchase difficult-to-value
assets from banks and financial institutions. However the concept
was lost in the immediate need to provide liquidity and stop banks
going bankrupt. Also there were problems over how to value and
purchase these opaque troubled assets in non-functioning markets.
The first $350 billion of TARP capital was used mostly to recapitalise
financial institutions, including the car finance industry ($255
billion in varying amounts to <120 banks, $40 billion in pref
shares of AIG, $20 billion to Federal reserve bank NY as equity
tranche of TALF, $20 billion to Citigroup, $14 billion to GM/Chrysler).
(ii) The Citigroup model
However, of all the first-round TARP money,
the $20 billion to Citigroup came closest to fulfilling the original
"bad bank" concept. For the $20 billion investment the
US Treasury and Federal Deposit Insurance Corporation (FDIC) received
$7 billion in pref stock with explicit participation in gains
and losses for $306 billion of ring-fenced troubled assets. The
potentially toxic assets remain on Citi's balance sheet and their
precise composition will be disclosed at some future date. So
this approach gets around the issue of having to value and disclose
the assets in question. In many ways though, the Citi model is
an imperfect solution. It lacks transparency and, in not removing
the assets from Citi's balance sheet, the capital relief is not
substantial or immediate, meaning that the bank is less likely
and less able to go out and write new loans. The bank is strengthened,
but the wider public policy objectives are not achieved.
(iii) TALF, term asset-backed securities loan
Not a "bad bank", but highly significant
as it marked a new form of intervention with government acting
as a buyer of last resort. The Fed was mandated to go into the
market and buy $600 billion of agency Mortgage Backed Securities
debt (MBS). The result was that agency MBS yields tightened significantly
and mortgage rates (closely correlated to MBS yields) have fallen
from 6% to 5.1% today (see graph). The remaining portion of the
TALF sees $20 billion of Fed equity being levered 10x to create
$200 billion of loans to buy AAA rated Asset backed securities
for newly originated auto/student/credit card/small and medium
company loans. As this portion of the TALF commences in February
2009, it should start to have a similar impact on consumer credit
pricing/credit availability in the US as it is already having
successfully on US mortgage rates.
Chart below shows falling US mortgage rates,
with TALF impact at end of period.
(iv) UBS model, similar to Citi except assets
moved off balance sheet into a Special Purpose Vehicle (SPV)
$60 billion of risk assets placed into an SPV
funded by a $54 billion 8 year loan from the Swiss National Bank
(SNB) at Libor+250bp and $6 billion of equity from UBS. This $6
billion of equity in the JV was effectively sold to SNB by UBS
for $1, and UBS had to raise a further CHF6 billion to restore
its capital ratios. There is no further recourse to UBS, but in
the event of any upside UBS stand to gain a share of profits.
This approach reduced UBS' hard-to-value assets from CHF55 billion
to <10 billion, with resultant capital relief, in one fell
* IMF Working Paper: http://www.imf.org/external/pubs/ft/wp/2008/wp08224.pdf
** Riksbank: http://www.riksbank.se/upload/1722/970829e.pd