Forecasting Bank Financials Methodology

Contacts
Jacques-Henri Gaulard Sam Theodore Pauline Lambert

[email protected] [email protected] [email protected]
Forecasting Bank Financials
Methodology
February 2014
Scope Ratings
Forecasting Bank Financials Methodology

February 2014 2 / 65

Contents
Executive Summary ............................................................................................................................................... 3
Forward-looking estimates in a rating context: why? ........................................................................................................... 3
Forward-looking estimates in a rating context: Which time frame should be used? ............................................................ 4
Forward-looking estimates in a rating context: What we learned ........................................................................................ 4
Choosing the appropriate metrics and the appropriate accounting system ............................................................ 5
Forecasting the balance sheet ............................................................................................................................... 7
Loans forecasts ................................................................................................................................................................... 7
Gross loans vs. net loans: introducing asset quality ............................................................................................................ 8
Deposits forecasts ............................................................................................................................................................... 9
Securities forecasts ........................................................................................................................................................... 11
Wholesale funding forecasts ............................................................................................................................................. 12
Putting this all together... ................................................................................................................................................... 14
Derivatives or the unfunded balance sheet ....................................................................................................................... 16
Other assets and liabilities ................................................................................................................................................. 18
Shareholders equity .......................................................................................................................................................... 19
Forecasting the Profit & Loss account ................................................................................................................. 23
The revenue block ............................................................................................................................................................. 23
Net interest income ......................................................................................................................................................... 23
Commissions & Fees ...................................................................................................................................................... 25
Trading gains .................................................................................................................................................................. 26
Net insurance income ..................................................................................................................................................... 26
Other income ................................................................................................................................................................... 27
The cost block ................................................................................................................................................................... 27
The cost of risk block ......................................................................................................................................................... 29
The below-the-line block ................................................................................................................................................. 33
The non-recurring items .................................................................................................................................................. 33
Taxes and minority interest ............................................................................................................................................. 36
Wrapping-up the forecasts: adjusting the balance sheet for earnings and distribution ......................................... 38
Forecasting the business lines ............................................................................................................................ 41
Divisional balance sheet 1.0 .............................................................................................................................................. 41
Divisional profitability forecasts ......................................................................................................................................... 42
dRCB .............................................................................................................................................................................. 42
fRCB ............................................................................................................................................................................... 43
WAM ............................................................................................................................................................................... 43
WIB ................................................................................................................................................................................. 46
Corporate centre ............................................................................................................................................................. 51
Putting it all together: art and science again ...................................................................................................................... 52
Regulatory forecasts ............................................................................................................................................ 54
Basel 3-compliant risk-weighted capital ratios ................................................................................................................... 54
Basel 3-compliant unweighted Leverage Ratio ................................................................................................................. 56
Net Stable Funding Ratio (NSFR) ..................................................................................................................................... 57
Computing the numerator: the available amount of stable funding ................................................................................. 57
Computing the denominator: the required stable funding for assets and on-balance sheet exposure ............................ 58
Liquidity Coverage Ratio (LCR) ......................................................................................................................................... 60
Calculation of the numerator and the denominator ......................................................................................................... 60
Scope Ratings


Executive Summary
This report looks at the determination and use of forward-looking estimates (or forecasts) in the ratings
assignments of Scope. As a measure of consistency, this report should be read in conjunction with Scopes
recently published Bank Rating Methodology. As such, the objective of this methodology is twofold: first, give
pointers towards ways to estimate the main metrics of the balance sheet and the P&L of a bank, and second,
show how specific metrics need to apply to the seven business model categories identified by Scope in its
methodology.
Forward-looking estimates in a rating context: why?
The objective of designing earnings and balance sheet forward-looking estimates is to improve the quality of the
analysis underpinning Scopes bank ratings. There are four main reasons as to why Scope uses estimates.
First, ratings are meant to be forward-looking, therefore they should be underpinned not only by past
quantitative data but also by forward-looking, carefully-chosen metrics.
Second, forecasting the financials of a bank forces an analyst to always remain on the ball, close to the
information flow communicated by the company on business, financial and regulatory matters. Such a
proactive attitude reduces the chances of a credit analyst missing something and therefore reduces the
possibility of catch-up multi-notch downgrades which may be hard to justify.
Overall, we believe estimates can make the rating more credible: indeed, the rating process is supported
by estimates on capital, funding, liquidity (and also earnings), which means that an analyst can justify with
much more accuracy the rationale behind any given rating. Being more grounded in reality and somehow
more quantified, the rating should be more accurate and therefore more reliable. It is also much less
static, as it projects an outlook several years out, and therefore fully justifies its title of forward-looking.
Part of the rating process will be based on the level that we expect some metrics to get to at a certain point
in the future. As explained in Scopes Bank Rating Methodology though, quantitative metrics are just one
aspect of the rating process and they should not substitute for a thorough qualitative analysis, which is an
indispensable complement to the whole rating process.
Lastly, it is important to note that Scopes forecasts will at all times be based on public information. Even
in the case where Scopes analysts teams are privy to confidential information communicated by the
company, our forecasts will always be based on our own scenario determined by public information. This
aspect is critical to Scopes banks ratings and is consistent with our Bank Rating Methodology.
This report gives a rather comprehensive framework for forecasting. The details are used to illustrate the
intellectual logic behind the forecasting work, but in practice it is only in the case of very complex and very large
banks that all the aspects of this methodology will effectively be used. A majority of banks in Europe and beyond
tend to be less diversified and many operate largely in a single country and in reasonably straightforward
deposit-taking and lending operations. As a result, for these institutions, an accounting forecast of the balance
sheet and profit and loss account is enough for the analysis.
As far as this accounting layer is concerned, our forecasting work will try and estimate:
The main balance sheet metrics around loans and deposits; financial instruments (including derivatives)
and wholesale funding; other assets and shareholders equity;
The main profit and loss account metrics around the revenue block; the cost block; the cost of risk block;
and the below-the-line block.
For more complex, multi-product, multi-business and multi-geography institutions, it is necessary to switch to a
more analytical level, and use the banks business line reporting to come up with reliable predictions. Lastly, for
both straightforward and complex banks, there is a mandatory regulatory layer that has to be included in the
forecast work considering the growing importance of this aspect in bank analysis.
Scope Ratings


Forward-looking estimates in a rating context: Which time frame should be used?
In the interest of keeping a sound level of detail while examining future financial metrics, we believe that
estimates for the current year plus the two following years are sufficient for the purpose of the rating analysis.
However, we believe that longer estimates (up to five years) could be helpful in one instance: assessing the
likelihood of success of a five-year business plan that would be communicated to the market in the course of an
investor day, for example.
This would represent an exception though. Scope is cautious on the absolute value of rating through the cycle,
as we made clear in our Bank Rating Methodology. The lengthy financial crisis which started in mid-2007
proved once again that apparently resistant structures can be easily washed away by cyclical trends.
Moreover, beyond a time range of two-three years, we believe that the risks attached to the forecasts diminish
their interest for the rating process, as the accuracy of the forecasts tends to materially decrease. The visibility
on a banks structure and business model, the interest rate cycle, market conditions and regulatory conditions
decreases over time.
Conversely, Scope believes it can proxy an adequate overview of a bank for two future years, everything else
being equal, to support the rating process.
Scope intends to amend its forecasts twice a year, after the H1 interim stage, and after the publication of the
annual report and the full year accounts. Scope will take exception to this rule if there is a material divergence to
estimates in the course of the Q1 and Q3 reporting or if a major corporate event occurs (acquisition, merger,
disposal, litigation, material profit warning, etc.), all potentially corresponding to a change in the credit narrative
of an institution.
Forward-looking estimates in a rating context: What we learned
Our comprehensive work on banks forecasts led us to the following conclusions:
Overall, qualitative aspects are as important as quantitative analysis.
There are many different assumptions to be made and getting all the numbers to simply add up (for example
making sure that a divisional P&L matches a consolidated group P&L) is no simple task. Paradoxically, the
increased complexity of bank analysis since the crisis has not diminished the problems: there has never been
more readily-available public disclosure, but this in turn means that there are more variables to analyse, and
therefore more opportunities to be wrong. Moreover, a lot of divisional numbers are restated and shuffled
around, making the ability to draw conclusions difficult over a long period.
Banks disclosure is far from perfect.
In a way, fair value accounting has made bank forecasting more difficult. The introduction of Trading/Available-
For-Sale (AFS)/Held-to-Maturity (HTM) has put the focus on the wrong thing by relegating the underlying
products (government bonds/corporate bonds/equities) in the background. Also, the quantity of pages available
in one annual report still makes some easy things difficult to find. For example, collecting the short-term debt of
a bank looking at YE 2012 reports is a complicated task while it should be a foregone conclusion.
Company guidance should be used selectively.
A good way to use what is called company guidance is to assess whether the guidance of the same company
in the past has been reliable or not. Having said that, there are areas where the official guidance (on Basel 3
metrics in particular) is the only way to forecast the relevant metric with a minimum of comfort.
Stress-testing is better than stress-forecasting.
In the past, it was frequent for analysts to presume the worst following difficult market conditions. Some
forecasts were probably too conservative. Stress-testing and examining the outcome of a balance sheet item or
a P&L line when changing parameters remain the best way to stress a companys financials without ruffling its
Scope Ratings


feathers. This said, as a rating agency, Scope is not bound by the same short-term dynamics as other market
participants. Quarterly financials are important but much less so than half-yearly financials, which in turn are
less important than the annual balance sheet and profit and loss account. Forecasting has a better chance of
being useful if it sticks to annual metrics, amended by the quarterly reports in the course of the year.
Why rating analysis is different from equity analysis.
Even if the tools used by Scope in its forecasting work are similar in nature to the craftsmanship of the equity
analysis, the objectives are different. Credit rating analysis does not navigate quarterly per share numbers with
the idea of staying close to a daily market valuation so as to identify trading opportunities. Nor does it have to
take quick views on the back of a constantly evolving stock price. On the contrary, the credit rating analyst
builds a forecast so as to identify weaknesses or uncertainties which in turn will have a direct impact on the
banks ability to meet its contractual financial commitments on a timely basis and in full as a going concern. So
clearly the focus is not on the next quarter, not even on the next half year, but much more on the capacity of the
bank to improve or secure its credit quality over a long period of time. The forecasting work will also help the
credit rating analyst to identify at an early stage the seeds of what could turn into significant credit problems in
future.
Keeping it simple.
This report gives a relatively comprehensive view of the forecasts process. Following the instructions by the
book is by no means a guarantee that the bank will deliver balance sheet and P&L metrics in line with
estimates. On the contrary, summarising the forecasts in 10-20 key metrics can be enough to have a solid view
of the institution being analysed. In a banking world that is getting ever more complex, it is tempting to get lost in
the detail. The European banking sector has been given for dead repeatedly over the last six to seven years.
Despite the financial crisis, macroeconomic problems, sovereign crises, material regulatory and policy changes
as well as public aid (at times), it is still here to tell the tale and in a much better financial condition than before
the crisis.
Choosing the appropriate metrics and the appropriate accounting system
We believe that forecast work has to be done at three different levels:
The accounting level of the balance sheet and the profit and loss account;
The analytical level of the business lines within one bank;
The regulatory level of the banking group as a whole.
Ideally, the analyst should reconcile the balance sheet and the P&L account of each division with the
consolidated P&L and balance sheet account of the consolidated banking group as a whole.
Most banks do this but with dramatic differences in the level of detail available. Therefore it is often not possible
to predict the balance sheet and the P&L account of a banking group as precisely as one would wish, as a lot of
information is missing from the business lines accounts.
Again, it is only for large and complex banks that Scope will use all the forecasting tools described in this
methodology. In a majority of cases, it is not necessary to run a very complex set of business lines forecasts.
All banks should be subject to the third layer of forecasting, related to regulatory and Basel 3-related metrics.
After identifying the breadth of forecasting tools that will be needed for each specific bank, the first complexity to
be solved is to choose the terminology of each item that needs estimated. Indeed, each bank is susceptible to
use a different accounting standard, even if international comparisons are much easier than they used to be.
Depending on the standard, the wording used to express the same accounting concept will be different.
Sometimes even, the same accounting concept will be computed differently.

Scope Ratings


The major accounting systems can be split in three categories:
International Financial Reporting Standards (IFRS) is an accounting standard used by the largest banks in
Europe. It has been implemented by 7000 companies in 25 countries in Europe in 2005. The application of
IFRS is required for consolidated financial statements of public companies that are listed in any EU
Member State.
United States Generally Accepted Accounting Principles (US GAAP) which can sometimes be used in
Europe on top of being the standard for the United States.
Lastly, for smaller institutions, generally unlisted, there is the possibility to use a Local GAAP to report
their accounts.
Since the majority of the banks to be rated by Scope are listed, they tend to report under IFRS, but Scope will
also rate companies reporting under US GAAP and local GAAPs.
The differences between US GAAP and IFRS are narrowing but they are still material and they can have an
influence on the forecasts: for example, US GAAP introduces the concept of VIEs (Variable Interest Entities)
whereby an entity which the bank does not own can still be fully consolidated. The equivalent concept under
IFRS on the other hand, SPEs (Special-Purposed Entities), have to be majority-owned by the bank to be
consolidated.
Another difference is that, even if IFRS and US GAAP are both historical-cost based accounting systems, the
methodology to calculate the fair value is more closely defined under US GAAP. Therefore accounts reported
under US GAAP are materially more fair-value based than the accounts reported under IFRS.
Lastly (and most importantly) the main difference between both accounting systems can be found in the area of
derivatives accounting. Under IFRS, a financial asset and a financial liability are offset only when there is both a
legally enforceable right to offset, and an intention to settle net or to settle both amounts simultaneously. Under
US GAAP, the rules are the same except for derivatives with the same counterparty and related collateral: these
may be offset against each other provided that they are subject to a master netting agreement, and certain
criteria are met.
The latter difference is the source of major asset size discrepancies between European banks and US banks.
This phenomenon therefore plays an important role in our forecasts of the leverage ratio, for example. The size
of the derivatives portfolio needs to be assessed carefully as it can trigger material differences in capital levels
that are just due to accounting differences.
Overall though, the concepts are similar, the activities are similar, and the financials of a bank in France can
easily be compared with one in Germany, one in the US or one in Japan. Therefore we will use a common
terminology covering all the major (but not all) components of the balance sheet and the P&L account.

Scope Ratings


Forecasting the balance sheet
To illustrate this section, we will use the simplified balance sheet of Bank XYZ. Bank XYZ has a very simple
business model of loans and deposits. The loans are fully matched by deposits. The bank has an ample capital
base (leverage ratio of more than 16%) that is kept in cash on-balance sheet. The balance sheet below is the
last published annual balance sheet of XYZ. We call that particular year Year Zero.
Figure 1: Balance sheet Bank XYZ (Year Zero)
Assets Liabilities
Cash 20 Total Deposits 100
Total Loans 100 Total Liabilities 100
Shareholders' Equity 20
Total Assets 120 Total Liabilities & Shareholders' Equity 120

Source: Scope Ratings
The core of the forecast work will be done on the loans and on the deposits. Scope will aim to be as accurate as
possible, and also conservative in forecasting, so that our ratings can absorb a sufficient degree of volatility with
respect to profitability or balance sheet metrics.
Loans forecasts
Loan growth is demonstrably a proxy of economic growth. So depending on this growth (which will be materially
different based on the geographic location of the bank) the growth rate of the loan book should be adjusted
accordingly. At times, it should be possible to break the loan book down in different sectors. The most common
split is the breakdown between retail and corporate lending. Another common split, within retail banking, is
between mortgage loans and consumer loans. Figure 2 puts this breakdown into perspective.
Figure 2: Possible lending breakdowns
O/W MORTGAGE LOANS O/W CONSUMER LOANS O/W SECTOR 1 O/W SECTOR 2 O/W
Total loans
O/W RETAIL LENDING O/W CORPORATE LENDING

While it is possible to fine-tune this breakdown, particularly on the corporate side, we do not feel this is a level of
detail we need to go into to establish forecasts (even if the sectorial breakdown of the banks loan book itself is
an important component of the rating process itself). Experience shows that, except for mergers, acquisitions or
disposals, the sectorial loan breakdown of a bank tends to remain broadly stable.
Another type of breakdown which is worth considering is the geographic breakdown as it shows the different
areas of growth. This breakdown can prove more useful than the sectorial breakdown to assess the potential
growth of the loan book but it is usually difficult to reconcile both.
Judgement is an important factor here: for a bank mostly present in one or at best two countries, the sectorial
breakdown will be the most useful to establish forecasts. For a multinational bank with a significant presence in
many markets, a forecast based on the weighted-average GDP growth of each country could be a better place
to start.
For a mono-country bank, a sectorial look is indispensable, and the split between corporate and consumer
loans will be key. In particular, for mortgage specialists, a thorough assessment of property prices, household
debt and Loan-to-Value (LTVs) will be important to forecast the loan book.
Scope Ratings


Another key factor is the maturity of the loan book. As such, a loan book can really be broken down in three
components:
The new loans that have been granted in the course of the year (and commonly known as the front book);
The loan outstanding that are in the normal course of their interest and capital payments (and commonly
known as the back book);
The loans that, in the particular year under review, are on the bulk of being repaid and/or refinanced (the
maturing loans).
Therefore at the end of one given year, the stock of loans at one given bank should be defined as follows:
Loan outstanding Year 1= Loan outstanding (or Back Book) Year Zero + Front Book Year 1 Maturing loans
Year 1.
Therefore one can see that the loan progression of a bank from one year to another can be really fine-tuned into
quite a bit of details. Unfortunately, public disclosure going to this degree of detail by business line is rather
infrequent except if the bank is active in few markets and geographies, in which such detailed forecasts are
possible.
Another aspect to take into account is the state specific sector support. In the course of some time periods,
the state will encourage banks to lend at a preferred rate to such and such industries: car loans, or first-home
buyer loans are typical areas where the state will subsidise the banks and retrocede to them the difference
between the nominal rate paid by the borrower and the cost of funding (plus a margin).
Last but not least, deleveraging is a factor in forecasting. For reasons linked to the necessity for banks to
strengthen their balance sheet, some of them will either reduce their lending effort or sell whole portfolios of
assets. In some cases, banks will reduce lending growth because the activity is expensive in liquidity terms.
This would be typically the case for consumer loans, funded with short-term borrowings and therefore expensive
for banks in light of their Liquidity Coverage Ratio (LCR) and Net Stable Funding Ratio (NSFR) calculations.
We are now ready to make a loan forecast on Bank XYZ.
Let us assume that Bank XYZ operates retail businesses in three to four low-growth mature economies, and
that management has given a cautious outlook for growth in these economies.
Let us also assume that Bank XYZ is just out of what has been a very difficult crisis and is not keen to go back
up the risk curve and that it will just refinance the loans arriving at maturity. We believe it is probably fair to
assume a flat loan portfolio this year (which we will call from now on the Current Year) versus last year (Year
Zero).
Improving asset quality and strengthened capital ratios should encourage XYZ to capitalise on its strength to
accelerate its growth and capture new clients from next year onwards. Its strong balance sheet should enable it
to compete on pricing in the hope of successfully cross-selling these clients throughout the firm. We therefore
believe it makes sense to assume that XYZ will grow faster than the economy in years one and two, with loan
growth of 2.5% and 3.5% respectively.
As one can see, the forecast process is certainly not only a function of objective measures. There is also a solid
dose of judgment, assessment of management targets and guidance, common sense, but also strategic
thinking: in light of Bank XYZs objectives, what is the best lending strategy considering the banks financial
constraints? And does this make sense versus the way the bank has delivered versus its targets before?
Gross loans vs. net loans: introducing asset quality
On balance sheet, the loans are usually reported net, i.e. net of loan loss reserves that are deducted from the
loan book to account for existing or future losses. What is reported is, in effect:
Scope Ratings


Net Loans = Gross Performing Loans + past due/impaired/non-performing loans Loan loss reserve
The level of past due/non-performing loans will clearly be dependent on the credit cycle and the underwriting
standards and there should be enough history at individual or central bank level for the analyst to have a solid
view on non-performing loans and loan-loss coverage over several credit cycles for one given bank.
The NPL (non-performing loan) forecast as well as the associated loan loss reserves should therefore be a function
of where the analyst considers one bank will be in one given stage of the credit cycle over the next two years.
When available, the analyst should take into account forbearance practices and their impact on loss-recognition
timing, as we make clear in Scopes Bank Rating Methodology, but this is not something that can be
generalised to the whole European banking sector yet because of lack of proper disclosure.
Deposits forecasts
Let us now switch to deposits: deposits are a more unified species than the loans. Overall, there are three
types of deposits. The first is demand (or sight) deposits representing the current account of households, where
money is deposited and withdrawn at no notice. Second there are time deposits (where depositors can only
withdraw money after a certain time has elapsed). Third there are savings deposits (paying a higher interest of
interest because the maturity is longer or because the withdrawal conditions are more stringent).
Deposits can also be split by categories of holders: retail deposits are considered more sticky, more stable,
even in the eye of the regulators. Corporate deposits are treasury instruments for companies; therefore they can
be very volatile as corporate deposits are a very competitive market. Regulators add to the breakdown by
distinguishing between covered deposits, non-covered preferred deposits and non-covered, non-preferred
deposits (please review our Bank Rating Methodology for more details).
Typically, because the retail/corporate breakdown of deposits can be difficult to obtain, deposits are forecast on the
basis of their maturity breakdown (which is also the accounting one): in other words demand, time and savings.
Four factors have a decisive impact on deposit forecasts:
Economic growth: even if less intuitive than for loans, the relationship between deposits and economic
growth is real: increasing economic growth increase capital expenditures, partly achieved through lending.
The capex create new employment therefore new salaries and then new deposits.
Inflation: the relationship between inflation and interest rates is well-known and can be mutually self-
sustained. In mature economies though, inflation rates have not been high enough so as to play an
important role in deposit pricing.
Interest rates levels have been key determinant on deposit pricing and most recently have been the key
driver behind the deposit growth at banks. Indeed, the recent financial crisis exposed the excessive
leverage level of many banks with regards to how much of their long-term loan book was funded by short-
term, volatile market resources. The recent strengthening of bank liquidity rules has encouraged banks to
increase the use of deposits in their funding (1) by shifting away from intragroup funding for their foreign
subsidiaries and (2) by entering very intense and sometimes unreasonable competition on interest rates to
attract depositors. Regulators in some countries have tried to bend this phenomenon. Overall though,
banks have been successful in strengthening their deposit base a way for banks to improve their loan to
deposit ratio was also to deleverage by reducing their loan book.
Lack of alternative for household savings. This last factor is more art than science but it reflects the
fact that despite the low economic growth most banks have experienced supernormal deposit growth driven
by customer reluctance to invest in equities.
Forecasting deposits will end up being significantly more complicated than loans: management guidance or
(even better) a quick look at the banks commercial websites can help assess the types of promotional offers
that can be offered on deposits. The analyst must therefore assess to which extent a bank has a better chance
Scope Ratings

February 2014 10 / 65

in capturing viable market shares and clearly assess the impact of this commercial effort on the interest paid
on deposits (a critical issue but one we will address when we speak about the net interest margin forecasts
later on). The management guidance can take several forms and can take the shape of a Loan-to-Deposit ratio
target for example. The analyst must in turn assess whether this target is reasonable or not.
After this effort is done, it is critical then to allocate the growth in-between sub-categories (demand, term, and
savings). And here the level and anticipation on interest rates will be essential. As interest rates go down, it is
likely investors/depositors are less concerned about leaving some more money on their current accounts as the
opportunity cost gets less important. As interest rates go up, there will be more movements on the terms and
savings side.
We note that the ultimate degree of maturity play in deposit terms for a household is quite often achieved
through life insurance products. They can be forecast at the same time as deposits but they will enter the very
specific bucket of technical provisions of insurance companies. The rationale for the forecast is exactly the
same as for deposits, but the accounting line is distinct, way down the balance sheet presentation reflecting
the less liquid aspect of this particular liability. Technical provisions in insurance terminology are defined as
intended to represent the current amount the insurance company would have to pay for an immediate transfer of
its obligations to a third party. In a way, the technical provisions are the counter-value of the total insurance
benefits and therefore are a good proxy for the life insurance product itself.
Mutual funds and assets under management (for institutional, retail and private clients) are all off-balance sheet
products as not liabilities of the bank itself. As such they can only be forecast at the business line levels of asset
management, wealth management, sometimes retail banking.
We are now in a situation where we can forecast the deposit base of Bank XYZ for the Current Year and the two
years after that. In light of the willingness of the bank to increase its liquidity, it has offered an exceptional 4.25%
yield on current accounts for new clients. This promotional offer is only valid for the first six months of the year.
XYZs management knows that this will have a negative impact on profitability, but is interested by the stock of
deposits first and foremost. We expect a more muted growth in the two years after that, slightly below the loan
growth. So we apply a deposit growth of 5%, 2% and 3% for the Current Year, Year 1 and Year 2 respectively.
We can now forecast Bank XYZs initial balance sheet for three time period, after the published balance sheet of
Year Zero.
Figure 3: Balance sheet Bank XYZ Current Year
Assets Liabilities
Cash 25 Customer deposits 105

Figure 4: Balance sheet Bank XYZ - Year 1
Assets Liabilities
Cash 24.6 Customer deposits 107

Scope Ratings

February 2014 11 / 65

Figure 5: Balance sheet Bank XYZ Year 2
Assets Liabilities
Cash 24.2 Customer deposits 110

It is important to note that, as loans and deposits do not grow at the same rate, the assets and liabilities should
not be balanced. However, since the towering principle of double-entry accounting lies in the equality of assets
and liabilities, we have assumed so far that the excess deposits generated by what we believe will be Bank
XYZs strategy will be reinvested in cash. This is not as simplistic as it sounds. When XYZs balance sheet gets
much more complex, a good way to match assets and liabilities that are not equal is to consider an increase or
a decrease of cash: a decrease of cash reflects the fact that the bank needs to allocate more money in granting
new loans, buying securities, contracting repo operations. Conversely an increase in cash means that so far the
bank has not really managed to find an appropriate usage for excess resources be it capital, debt or deposits.
So basically the available cash of the bank is paradoxically a metric that never gets to be forecast but that tends
to be an outcome of the forward-looking strategy of the bank, as expressed by the analysts forecast even if
regulatory constraints have recently turned cash balances into more important and less predictable components
of the balance sheet.
Securities forecasts
In the area of markets and securities, the accounting norms are not very helpful. Indeed, securities, under IFRS
or US GAAP, are part of an item labelled financial instruments. Financial instruments are split between cash
instruments and derivatives instruments.
Financial instruments are not split by product, as would intuitively made sense but by intention and relative
liquidity. So for example,
Trading assets are instruments where the variation in value (either by sale of revaluation of the asset) is
channelled through the P&L.
Available-for-sale (AFS) assets are instruments where the change in value is channelled through
shareholders equity.
Held-to-Maturity (HTM) assets are instruments which the bank commits to keep in its books until maturity at
historical costs.
To do a proper forecasting job, the analyst has to retrieve debt instruments and equity instruments from the
three components and apply some growth rates accordingly.
Within the debt securities category, it will be helpful to split government bonds from other bonds. Even if in
many instances the sovereign component of banks balance sheets is nowhere nearly as risk free as it used
to be considered, it is still a vital component to assess the carry-trade activities of a bank, i.e. the use of short-
term market/interbank resources to fund the debt of one given state.
These activities present two risks and one opportunity for the bank. The opportunity is to secure a strong spread
reflecting the difference between the yield on the government bond and the interest paid on the short-
term/interbank resource.
The first risk is the counterparty risk of some sovereigns, which could lead to depressed bond prices if the credit
risk of these sovereigns was perceived by the market to deteriorate. The second risk is over-dependency on
Scope Ratings

February 2014 12 / 65

short-term Central Bank resources, preventing an institution from funding itself properly through diversified
market and customer funding.
In this context, the role of the credit analysts forecast is to assess the timeframe from which the bank will have
made enough progress in its liquidity and its balance sheet management to significantly reduce its dependency
on Central Bank funding.
Typically, forecasting equities and bonds reflect a relationship between volumes and prices.
For equities, a quick look at overall stock price performances in the major countries in which a bank operates
can give a good idea of the buoyancy of the equities market and therefore of the relative situation of the equities
portfolio. The analyst will have to assume that the bank may well have sold some equities into a strong market
so that the performance may not be fully reflected in the balance sheet growth of the equities portfolio.
The stock of corporate bonds will be essentially triggered by interest rates levels and by credit risk since they
will have a direct impact on the valuation of the bond portfolio both government and corporate. Regulation
around proprietary trading and the capital cost of holding too many bonds in inventory are also factors that credit
analysts have to keep in mind when forecasting bond holdings of banks.
As for government bonds, the perception of any given sovereign, its rating and the level of interest rates will be
key drivers to the forecasts.
Wholesale funding forecasts
Similar to the clustered analysis of loans and deposits, we treat securities and wholesale funding in the same
bucket. Logically, and in order to reduce risks, a bank should fund its financial assets by wholesale funding.
However, part of the wholesale funding will be used to fund the loan book. Even if this phenomenon is now less
material due to more restrictive regulation, match-funded loans and deposits are less frequent and there are
countries in which the funding of the loan book is still to some extent dependent on wholesale funds.
What do Scope ratings define as wholesale funds?
A quick look at our Bank Rating Methodology defines wholesale funds as interbank deposits + debt.
Interbank deposits would include traditional demand and term deposits from banks but also repo operations. We
believe that in some banking systems, collateralised lending from Central banks is included in the repo line.
Debt would include short-term debt instruments such as certificates of deposits plus long-term senior debt, but
also all the subordinated range of debt.
Some elements accounted for as own funds would have to be restated as debt: this would be old-style
preferred shares and other hybrids.
Repo/Interbank:
One of the most active short-term wholesale markets is the repo market: A repo is an operation where a bank
sells securities against cash, but is committed to buy the securities again at a given time in the future. For a
bank, selling securities like this is equivalent to lending the security (borrowing the money) even if the term is
misleading since the property titles are transferred to the borrower in the course of the operation.
Conversely, a reverse repo is the operation where the bank actually buys the security against the cash, which is
then sitting with a counterparty. In this situation, the bank is lending the money and borrowing the security (even
if all the property titles of the securities have been transferred to the security borrower). Because the bank owns
the security, the reverse repo sits on the asset side of a banks balance sheet.
Since the crisis, the interbank market which was very buoyant has declined significantly so that the bulk of
interbank operations tend to be collateralized thence the importance of the repo market.
Scope Ratings

February 2014 13 / 65

CDs:
Certificates of Deposits (CDs) and Commercial Paper (CPs) are short-term (typically less than a year) market
funds. They have been for a long time used by banks as a deposit substitute when the loan book was growing
faster than the deposit base. To be sure, as a market resource, CDs are more expensive than deposits but they
proved to be very liquid up to the point when the crisis occurred. Indeed, a lot CDs were bought by money
market funds, meaning than in many ways banks were keeping the control of this product since banks were
sponsoring the funds themselves. With the development of capital markets, banks started to issue CDs in non-
domestic currencies which enabled them to find funding beyond their traditional domestic market. The flipside is
that they became dependent on foreign funding sources and there have been cases where institutional CD
buyers decided to brutally withdraw from certain currencies or certain countries. This has forced banks to find
alternative funding in a hurry. With signs of economic recovery in the Euro area (EA), US money market mutual
funds are now re-investing in European banks CDs.
Long-term debt:
Following the crisis and increased efforts by the regulator to match-fund assets and liabilities (through the
introduction of new metrics such as the NSFR Net Stable Funding Ratio), banks have given more and more
importance to long-term funding. The recent initiatives on recovery and resolution are also forcing banks to
increase their level of bailinable debt as a percentage of their total liabilities. LT funding has become so
important in conditions of limited liquidity for senior unsecured resources that banks had to develop the covered
bonds market and allocate dedicated pools of assets to specific bond issues so as to secure interest from
investors.
Capital instruments:
Capital instruments, in theory, should not be part of the wholesale funds. But as the new recovery and resolution
regimes are increasing the type and number of instruments that are effectively loss-absorbing, the line between
debt and equity is more and more blurred.
Regulatory capital instruments such as Tier 2 capital and Alternative Tier 1 (AT1) instruments will have a
growing importance as the regulators force banks to raise more bailinable/loss-absorbing debt.
We believe that there are two drivers to wholesale funding forecasts:
1. Economic and market activity, that will somewhat dictate the funding of the loan book;
2. Regulatory trends, which will force to reallocate funds from short-term and volatile towards long-term
debt, either unsecured or secured, and towards capital instruments compatible with new capital and
resolution and recovery regulation.
While CDs and CP will still be looked after for filling the gap between loans and deposits, and while repos will
always be important, banks will have to increase the weight of long-term funding (through secured and
unsecured debt) to fulfil the need of the NSFR ratio and to abide by the new recovery and resolution regimes.
Therefore, these constraints will have to feed the analysts forecasting priorities, on top of the markets and
lending activities occurring on the asset side.

Scope Ratings

February 2014 14 / 65

Putting this all together
We are now in a position where we can afford to add complexity to XYZs balance sheet.
Figure 6: Balance sheet Bank XYZ Year Zero
Assets Liabilities
Due f rom Banks (incl. Reverse repos) 20 Due to Bank (incl. Repos) 80
Government Bonds
85 CDs 95
Other Bonds 65 Long-term senior debt 15
Equities
30 subordinated debt 10

Assets Liabilities
Cash 25.0 Customer deposits 105.0
Due f rom Banks (incl. Reverse repos) 20.0 Due to Bank (incl. Repos) 72.0
Government Bonds
72.3 CDs 80.8
Other Bonds 58.5 Long-term senior debt 18.0
Equities
30.0 subordinated debt 10.0
Total Loans 100.0 Total Liabilities 285.8
Shareholders' Equity 20.0
Total Assets 305.8 Total Liabilities & Shareholders' Equity 305.8

Assets Liabilities
Government Bonds
72.3 CDs 80.8
Equities

Scope Ratings

February 2014 15 / 65

Assets Liabilities
Government Bonds
75.0 CDs 75.0
Equities

As we can see in the figures above, we have added the different securities on the asset side, and the wholesale
funds on the liability side. The interbank deposits (including repos and reverse repos) appear as Due from
banks on the asset side and Due to banks on the liability side.
With the growth imposed to XYZ, the bank is now much more leveraged than previously, and the loan book is
not the most important component of the balance sheet anymore.
As Figure 6 demonstrates, the bank has a heavily negative net interbank position (meaning that XYZ has more
interbank liabilities than interbank assets). XYZ also posts an important securities portfolio (180 in total as of
Year Zero, so about 56% of the banks total assets).
We also note that the dependency of the bank on short term funding is quite important (close to 60% of total
liabilities).
As far as forecasts are concerned, the Current Year is showing that XYZ is trying to reduce its bond portfolio on
the basis of pressure on interest rates so we are expecting a 15% decrease in government bonds for the
Current Year, and a 10% decrease in the other bond portfolio. We assumed equities unchanged.
For Years 1 and 2, we appreciate the fact that XYZ does not need to alter the composition of its bond portfolio
(or only at the margin). Considering likely buoyant equity markets and considering that the performance of
equity markets should go beyond its historical averages, we assume that XYZ would grow the level of its equity
portfolio by 7% in Year 1 and +10% in Year 2.
On the funding side, XYZ is aware that it needs to rebalance its funding structure towards more long-term
products. Therefore we have assumed that for the Current Year, XYZ would decrease its interbank liabilities by
10%, and its CDs outstanding by 15%. We assume a 20% increase in senior debt: this seems a lot (and the
forecasts have to take into account the capacity for the market to absorb XYZs long term debt issues) but
considering the low initial level of senior debt in Year Zero, we do not believe that this amount is too demanding.
For debt forecasts, it is always interesting to refer to the debt maturity of a given bank. It enables the analyst to
calibrate more precisely the amount of debt that has to be refinanced. For long-term debt maturing within the
next twelve months, XYZ may choose to refinance it with LT debt or use more short-term instruments,
depending on the level of activity.
As for subordinated debt, we assume that these are products that will not be eligible for either AT1 or Tier 2
capital under the new regulatory regimes, and that the bank will exchange or call them (even if these products
can be used as bailinable debt). Therefore we have assumed that XYZ would reduce them by -20% and -25% in
Year 1 and Year 2 respectively.
Scope Ratings

February 2014 16 / 65

Following this process a reality check will be performed. Comparing the Year Zero balance sheet with the
forecast Year 2 balance sheet, our forecasts look reasonable. Deposits increase from 31% to 36% of total
assets; short-term funding (interbank, repos and CDs) decline from 55% to 48%, and long-term funding
increases from 4.7% to 7.5%.
These forecasts seem realistic enough and reflect reasonable efforts by XYZ management to restructure the
balance sheet while demonstrating at the same time that banks balance sheet structures are quite static and
difficult to adjust in a short time period.
Derivatives or the unfunded balance sheet
Derivatives are by definition financial instruments. Banks can use them for hedging but the vast majority of
these derivatives are held for business/trading purposes. Therefore, derivatives are reported in banks balance
sheet at fair value. Derivatives represent rights (assets) or obligations (liabilities) that meet the definition of an
asset or a liability therefore they should be reported in financial statements.
What makes them specific is the fact that:
They are derivatives of an instrument so they are based on one or several underlyings;
They require little payment at the outset of the contract;
They permit net settlement.
The latter specificity implies that profits and losses are realised without taking possession or getting rid of the
underlying asset this is why derivatives are considered as being unfunded.
Generally, banks disclose the underlying assets that derivatives apply to: interest rates, credit, currencies, and
equities so essentially other market instruments. Derivatives forecasts will therefore be able to proxy the
trends identified by the analyst on underlying bonds or equities. Interest rates derivatives (swaps) will often
make the bulk of a derivatives portfolio.
In the past, derivatives outstanding often had a life of their own. We believe that with (1) the gradual
disappearance of proprietary trading desks throughout banks due to more restrictive regulation and (2) the
increasing reliance on Central Counterparty Clearing Houses (CCPs) for OTC derivatives, derivatives are likely
to follow more closely the trend of a banks business. Therefore it seems logical that derivatives forecasts
should end up being a derivative of our loans and securities forecasts.
It is interesting to note that in the course of the recent deleveraging that followed the financial crisis, the
derivatives line has been one of the most reduced balance sheet items. Banks use different techniques, such as
derivatives compression, to reduce the volume of outstanding derivatives with frequent counterparties.
Sometimes they do not renew specific derivatives trades, which are therefore unfolding and they subsequently
disappear from the balance sheet. .
Assuming a bank would decide to let its derivatives exposure unchanged, the reported volume on balance sheet
would ultimately depend on interest rates movements, since interest rates-related derivatives (Swaps, CDS) are
the most customary.
If we add Derivatives to XYZ, and if we assume that the components of these derivatives are representative of
XYZs current asset structure and its expected growth over the years, then the banks balance sheet should look
as follows:
Scope Ratings

February 2014 17 / 65

Assets Liabilities
Government Bonds
85 CDs 95
Equities
Total Loans 100 Derivatives (obligations) 85
Derivatives (rights)
80 Total Liabilities 385

Assets Liabilities
Government Bonds
72.3 CDs 80.8
Equities
Total Loans 100.0 Derivatives (obligations) 79.2
74.5 Total Liabilities 364.9

Assets Liabilities
Government Bonds
72.3 CDs 80.8
Equities

Scope Ratings

February 2014 18 / 65

Assets Liabilities
Government Bonds
75.0 CDs 75.0
Equities

So if the derivatives book of XYZ follows its asset composition more closely, then the weighted average growth
of the book (based on the assumed growth rates of loans and securities) should be -6.875% in the Current Year
(versus Year Zero), +1.8% in Year 1 and +4.2% in Year 2.
We have assumed that the net derivatives position of XYZ was negative (i.e. the bank has more obligations than
rights).
Other assets and liabilities b
We now need to spend some time on other assets and liabilities, all of which being far from negligible.
Current and deferred tax assets (DTA) and liabilities:
This position grew in importance when it became clear that Basel 3 was unlikely to recognise the DTAs linked to
future profitability (namely the tax loss carry-forwards) and would cap the recognition of DTAs linked to time
difference. Obviously, the analyst should assume a steady reduction of Tax-Loss Carry Forwards (TLCFs) as
these TLFCs should be assumed to be deducted from core equity Tier 1 capital under the new Basel 3
regulations. As a result of this, DTAs are usually forecast at regulatory level, when the analyst has to estimate
the level of Basel 3 capital levels for a given institution. An exception is when the government of a country
explicitly guarantees the banks tax loss carry forwards, specifically to help banks avoid the Basel 3 capital
deduction.
Insurance assets & technical provisions:
The disaggregation of insurance financial assets from banking-related assets can be difficult to find in a banks
annual report. Therefore, the only way to reflect the weight of insurance operations is through technical
provisions. In terms of pure forecasting skills, technical provisions will be treated the same way as securities.
Without enough information, the analyst will consider the weighted-average growth of the securities of the bank
and will apply the same percentage to the technical provisions. It is also important not to forget that, in some
markets where bancassurance is particularly developed, an insurance product is simply another type of savings
deposit. In these specific markets, it makes sense to forecast life insurance products as savings deposits.
Provisions:
Theoretically, as we have seen in the loan forecasts section, specific provisions against loan losses are
deducted from the asset side. There are nonetheless provisions appearing on the liability side, and they usually
reflect two series of contingent liabilities:
On the one hand, provisions for employee benefits. For these, we pay particular attention to the gap
between the value of the assets underlying pension payments and the value of the obligations towards
employees.
Scope Ratings

February 2014 19 / 65

On the other hand, provisions for litigation. Litigation charges can have a very significant impact on a
banks profitability, as been recently demonstrated in the US and in Europe. Therefore, the proper
assessment of a banks litigation cases, a proper assessment of their outcome and their financial impact
are a growingly important part of banks financial analysis and forecast.
Equity-accounted investments:
This line reflects minority equity stakes in which the bank has a significant influence. This is presumed to be
the case when the ownership is 20% or more. The investment is initially recorded at cost and is subsequently
adjusted to reflect the investors share of the net profit or loss of the associate. This line will usually be fairly
stable over time (safe for the annual adjustment in the associates net profits) except when a fully-consolidated
equity stake gets partly sold, meaning that a bank ends up with a minority stake rather than a majority stake, for
example. This factor is likely to boost the associate line. Conversely a bank can decide to reduce a stake that
was previously equity-accounted - the idea being to avoid too important a capital deduction of that stake under
Basel 3. If enough of the stake is sold, the shares are deconsolidated and reclassified as AFS assets, triggering
a fall in the equity-accounted accounting line.
Intangible assets & Goodwill:
Operating intangibles (such as software expenses) are to be amortized and they should be distinct from
goodwill. The portion of goodwill that is not identifiable as PPA (or Purchase Price Allocation) is not to be
amortized but submitted to yearly impairment tests, validating or invalidating its valuation. Goodwill is generally
assumed stable except if the fair values of comparable companies tend to go down consistently, in which case
the analyst has to consider an impairment. We come back on PPA later in this methodology.
Fixed assets:
These are lands, buildings, equipment and furniture, sometimes to be depreciated. Ageing equipment has to be
replaced; therefore the forecast insight is pretty limited.
Other assets & liabilities:
At any given time, banks hold vast amounts of prepaid expenses and accrued income which are of no particular
analytical importance. As a result, this inventory of various metrics is sometimes used as a balancing item for
the balance sheet even if the cash balance is our preferred item for such a purpose. To be thorough, the
overall balance sheet growth averages could be applied to these items but we do not view them as critical.
Shareholders equity
The main challenge will be a proper identification of the components of shareholders equity. Under IFRS, the
own funds line can include a vast proportion of subordinated debt, preferred shares and other old style
capital instruments that have nothing to do with the proper loss-absorbing common equity that forms the last
line of defence of a banks credit.
To properly identify this line, we refer to the statement of changes in shareholders equity. There we make the
difference between:
Shareholders equity, group share;
Other Comprehensive income;
And Minority interests.
Shareholders equity should be understood in its strictest sense i.e. the sum of:
Share capital + issue premium + accumulated retained earnings + non-distributed income for the year.
Other comprehensive income (OCI) is really the accumulation of changes in the value of assets and liabilities
recognised directly in equity. Typically, this is where the variations of value on AFS assets will be recognized.
Scope Ratings

February 2014 20 / 65

Another big line in this category is the change in value of assets and liabilities triggered by foreign exchange
movements. This can be quite big for companies deciding not to hedge their foreign exchange risk.
Overall, the OCI is extremely difficult to forecast as it is a balance sheet reflection of all the market movements
that have affected a bank in the course of one given period.
As a result, the main difference in equity forecasts from one year to another is the net profit, minus the planned
distribution for the year. The forecasts will also reflect, as and when needed, capital increases and capital
reductions.
Assume a banks capital is divided in 1,000 shares with a nominal of 20. The bank decides to raise a sum of
2,000, at a price of 100 (that happens to be close to the share price). The way to reflect this in the forecasts is
to:
1/ Raise share capital by 400: nominal of 20 x number of new shares issued (2000/100, so 20);
2/ Raise issue premium by 1,600: issue premium of 80 (issue price of 100 - nominal of 20) x number of new
shares issued (20).
Lastly, convertibles and warrants may be included in equity depending on the conversion feature (mandatory
convertible bonds for example).
Assets Liabilities
Government Bonds
85 CDs 95
Equities
80 Current & def erred tax liabilities 3
Current & def erred tax assets 20 Technical provisions 20
Other assets 5 Other provisions 10
Equity-accounted investments
2 Other liabilities 5
Intangible assets 2 Total Liabilities 423
Goodwill
7 Shareholders' Equity 20
Fixed assets 2
Total Assets
443 Total Liabilities & Shareholders' Equity 443

Scope Ratings

February 2014 21 / 65

Assets Liabilities
Government Bonds
72.3 CDs 80.8
Equities
74.5 Current & def erred tax liabilities 3.0
Current & def erred tax assets 16.0 Technical provisions 17.4
Other assets 5.0 Other provisions 10.0
2.0 Other liabilities 5.0
Intangible assets 2.0 Total Liabilities 400.3
Goodwill
7.0 Shareholders' Equity 20.0
Fixed assets 2.0
Total Assets
420.3 Total Liabilities & Shareholders' Equity 420.3

Assets Liabilities
Government Bonds
72.3 CDs 80.8
Equities
Goodwill
Fixed assets 2.0
Total Assets

Scope Ratings

February 2014 22 / 65

Assets Liabilities
Government Bonds
75.0 CDs 75.0
Equities
79.0 Current & deferred tax liabilities 3.0
Goodwill
Fixed assets 2.0
Total Assets

We can now disclose a reasonably detailed balance sheet forecast for Bank XYZ, in Figures 14 to 17.
The balance sheet of XYZ in Year Zero is now complete.
We have assumed that XYZ had a small insurance operation (justifying technical provisions of 20), some
contingent liabilities for a legal case (provisioned up to 10).
Following past losses, XYZ also recognizes tax loss carry-forwards of 20. A recent acquisition justifies a
goodwill of 7.
With regards to forecasts for the Current Year, Year 1 and Year 2 (Figures 15 to 17), a lot of these items are
likely to remain flat (except if an impairment is recognized on the goodwill), except for:
DTAs related to Tax Loss Carry Forwards (TLCFs), that are subject to deduction under Basel 3 in XYZs
country, and that XYZ decides to use aggressively to ensure that these deferred tax assets are gone within
five years we have decided to depreciate the amount on a straight line basis accordingly;
And the technical provisions, where we have proxied the trends we forecast for the bond portfolio of XYZ
(both government and other). In this context, we expect XYZs stock of technical provisions to decrease by
12.8% in the Current Year, before stabilising in Year 1 and Year 2.
We have assumed flat legal provisions, knowing that these can go up and down depending on the outcome of
some legal cases and the reassessment of claims on others.
Please note that we did not forecast the shareholders equity of XYZ. This is due to the fact that this all-
important line can only be forecast while net profit and net distribution for the year have been forecast. This is
the next section of this methodology.

Scope Ratings

February 2014 23 / 65

Forecasting the Profit & Loss account
When forecasting the P&L of a bank, it is helpful to disaggregate it in four different blocks:
The revenue block;
The cost block;
The cost of risk block;
And the below-the-line block. This particular block refers to non-recurring items that can be included in the
three blocks above. Usually banks communicate the amounts that are usually deemed as non-recurring, so
that they can be easily restated by market participants. These one-off items are said to be below the line
so below the pre-tax profit line. They are usually not forecast because, as one-off, they are generally
unpredictable, unless the company guides specifically for them.
The revenue block
There is no particular common denomination to address a banks turnover. Some banks use revenues, others
gross operating income or Total operating income...some use net banking income. Other simply income.
We will stick to the expression used in our Bank Rating Methodology: Revenues.
Revenues can be defined as follows:
Revenues = Net interest income + net commissions & fees + trading income + net insurance income + other
income.
Net interest income
The net interest income is the difference between interest income and interest expense.
Interest income is received on loans, financial instruments, interbank deposits.
Interest expense is paid on debt, deposits, interbank liabilities and financial instruments.
There are two ways to forecast the net interest income.
Method #1: determining net interest income through net interest margin (in %).
We define Net interest margin (NIM) as net interest income divided by interest-earnings assets.
In turn, interest-earning assets can easily be traced in our balance sheet forecasts above. It is the combination
of cash & cash equivalent, interbank assets (including reverse repos) and interest-earnings securities held
(bonds and other debt instruments).
As a result, if a bank is straightforward enough to be a pure domestic retail player (i.e. with interest earnings
assets being essentially loans), it can be fairly easy to proxy what its net interest margin will be. The problem
with that methodology is that it is not only simple but also simplistic as it takes no account of the cost of funding
of the bank.
Method #2: determining net interest income through customer spread (in %).
We prefer this method as (1) it enables us to issue forecasts on both interest income and interest expense (and
not just the difference between both lines) and (2) it also allows us to consider the impact of the future direction
of interest rates in two dimensions: the revenue dimension and the expense dimension.
Customer spreads are defined as:
(Interest income/Interest earnings assets) (Interest expense/interest-bearing liabilities)
Scope Ratings

February 2014 24 / 65

Or,
Customer spread = Yield on earnings assets minus cost of interest-bearing liabilities.
Banks net interest income will evolve differently depending on interest rates expectations. Very simply put, net
interest income trends will depend on which, between loans and deposits, will reprice faster considering a given
move in interest rates (assuming all moves are parallel along the yield curve).
Basically, a bank will benefit from an interest rate rise if its loan book is priced based on variable interest rates
loans (and interest-earning assets) while the yield on its deposits is fixed. In this case, the customer net interest
margin of the bank is likely to go up.
Conversely, if a banks loan book is primarily fixed rates, it is likely that the funding of the bank will reprice more
quickly than its assets, meaning that the customer spread will go down if interest rates go up.
Forecasting by using customer spread is therefore slightly more precise than simply using the net interest
margin.
Net interest income analysis is nonetheless made complicated by two separate factors:
Hedging policies. To be less dependent on interest rates variations, banks can choose to hedge all or part
of their deposit base so as to crystallise a margin over a fixed period of time (typically five to six years).
The advantage is that the NIM shows little volatility over that period of time, which means that the bank can
concentrate on the more immediately commercial aspects of its business. The drawback is that these
hedges are not eternal and have to be repriced. They also sometimes have to be repriced on the basis of a
yield curve that has nothing to do with the yield curve that was prevalent when the initial hedge was put in
place, five of six years earlier. This phenomenon can create material revenue volatility. It is also
compounded by potential hedging on the asset side, when banks decide to turn fixed rate loans into
floating rates loans (and vice versa) through swaps;
The fact that interest income and interest expense cover non-NIM-related business. A significant part
of the interest expense of a bank is usually there as a carrying cost for the trading portfolio. As a result, the
trading income as reported by the accounts is a gross income, as only income from trading is recorded
there (without the corresponding carrying cost, accounted for as interest expense). Also, a large part of the
interest income is not directly linked to banking operations (as associated with financial instruments and
interbank operations). Therefore a notion like customer spread is much wider than just a spread between
customer loans and customer deposits. It really covers the difference in yield between the return on
interest-earnings assets and the cost of interest-bearing liabilities on a bank-wide basis and loans are
only a part of banks interest earning assets, the same way that deposits are only one aspect of interest-
bearing liabilities.
As a result of the above, a net interest margin forecast can be only be considered accurate for pure retail banks.
A NIM forecast can also work for more complex banks, assuming interest income and interest expense are
broken down by product (loans, deposits, etc.) in the accounts. .
Let us try to proxy a net interest income forecast for XYZ based on the final balance sheets appearing on
Figures 14 to 17.
We will assume that in Year Zero, XYZ reported a yield on interest earning-assets of 3.5% and a cost of
interest-bearing liabilities of 2%. From what we know of Bank XYZ, the company sustained no loan growth this
Current Year and considering that interest rates have remained overall flat, it probably makes sense to also
assume flat yield on interest-earnings assets of 3.5% for the Current Year. As interest rates are expected to go
up we would expect XYZs yield on interest-earning asset to creep up as well, so we increase the yield to 3.6%
and 3.7% in Year 1 and Year 2 respectively.
Scope Ratings

February 2014 25 / 65

As for the cost of interest-bearing liabilities, we know that XYZs management made a significant effort on
deposit market shares in the Current Year, which would warrant a significant increase from 2% to 2.25%. As this
marketing effort subsides, we would not expect XYZ to reprice its deposit base significantly. We would therefore
maintain the cost of interest-bearing liabilities to 2.25% in Year 1 and in Year 2 in the context of rising interest
rates.
Basically the net interest margin will be a function of the growth rate of loans (and other interest-earning assets)
and deposits (and other interest-bearing liabilities), as well as interest rates expectations. As a result of our
calculation above, we can infer the net interest margin of XYZ on a three-year period.
Figure 18: Inferred net interest margin of Bank XYZ (%)
!"#$%
!"&&%
!"'&%
!"##%
!"&(%
!"&)%
!"'(%
!"')%
!"#(%
!"#)%
!")(%
*+,- ( ./--+01 *+,- ! *+,- &

As can be shown in the chart above, market share gains on deposits imply a sharp fall in XYZs net interest
margin between Year Zero and the Current Year. A faster repricing of the loan book in a rapidly rising interest-
rates environment help XYZ restore its NIM by Year 2, almost back to where it was in Year Zero.
Commissions & Fees
The second part of the revenue block is made of fees & commissions. As for interest income, fees &
commissions are the product of a difference between commissions received and commissions paid.
However as a significant part of commissions are not asset-based, it is infrequent for an analyst to forecast both
commissions received and commissions paid. The forecast work in the P&L is limited to net commissions only
(i.e. commissions received minus commissions paid). Commissions paid reflect monies paid by the bank to
advisors on its debt and equity issues, for example. Under IFRS, commissions paid also include monies paid to
third-parties for business introductions (even if this particular item is recognized as an expense under US
GAAP).
Commissions are received on means of payment and on loans, but also from asset management and private
banking activities, capital markets operations and advisory business. Commissions also include brokerage fees,
etc.
In terms of forecasts and versus the net interest income, commissions have to be expressed as a growth
percentage versus the prior year. Because fees and commissions cover a multiplicity of businesses (there is
nothing in common between a processing fee on a mortgage loan and a management fee perceived on a
mutual fund) it is difficult to forecast them with a reasonable degree of confidence. In this respect, business line
revenue forecasts will give a greater degree of comfort than the commission line itself, which is linked to a vast
array of businesses.
Scope Ratings

February 2014 26 / 65

While we do forecast commission and fee income, this line will inherently be vaguer. For example, from a
reasonably depressed level of fees & commissions in Year Zero, XYZ is not experiencing client business to
grow significantly. Only in Year 2 does XYZ expect a modest recovery in commission income of 5% p.a...
Trading gains
Capital market revenues (or trading gains) are even more volatile and unpredictable than commissions. In this
domain, more detailed inputs from the business lines will be a much better base for forecasting than just using
the reported P&L number which, as we pointed out in the section on net interest income, is not economically
representative of the effective trading gains, as the cost of funding the trading portfolio appears at interest
expense level.
Trading gains is the commonly-accepted term used to refer to trading-related income. However, we have seen
that quite a chunk of capital markets income can be included in commissions as well.
As a result, trading gains really refer to capital gains, on two categories of products:
Realised and unrealised capital gains on assets held for trading (fixed income instruments, equity
instruments, F/X instruments, commodities instruments, and derivatives). Because these assets are held
for trading, they need to be valued on a mark-to-market basis, so the gains will go through the P&L even if
they are not realised;
Realised gains only on AFS securities. Unrealised gains on AFS are recorded as Other Comprehensive
Income (OCI). Once the AFS instruments are sold, the capital gain is transferred from OCI to P&L.
Overall, there is a significant volatility that is attached to trading gains making them, as such, very difficult to
predict. The task is made easier at business line level, when the bank discloses the breakdown of capital
markets revenues. We therefore feel more comfortable using the level of the business line to make forecasts on
capital markets activities.
However, determining the direction of trading gains can be helped by focusing on major market indices, credit
spreads, and any daily market information helping identify the trends that are likely to impact a banks trading
gains line even if that line gives little to no detail about the specific business model of a bank in investment
banking.
Using XYZ as an example, from a rather sustained level in Year Zero, we expect trading gains to remain flat in
the Current Year (the good performance in the course of the first half being negated by less favourable H2
market conditions). Year 1 should look more like the second half of the Current Year so we are expecting a 20%
decline in trading gains, followed by a muted 5% recovery in Year 2, driven by recovering equity markets.
Net insurance income
For some banks, insurance operations can represent an important portion of their business. The net insurance
income represents the difference between income from insurance activities minus claims and benefits
expenses.
Net insurance income includes gross premium written (on the income side), movements in technical provisions
(on the income or expense side), claims & benefits paid (on the expense side), surrenders and maturities (on
the expense side).
As is the case for trading gains, the presentation of insurance activities under traditional bank accounting is far
from ideal. We will tend to rely again mostly on business line disclosure to make forecasts.
Insurance is a complex business, whose full assessment may go beyond the competence of a bank analyst, but
as a way to proxy the revenue trends in insurance, a method consisting in applying a return on the technical
reserves is a good way to start.
Scope Ratings

February 2014 27 / 65

For the purpose of this exercise, and from a starting point of 2.5%, we assume a flat return for the insurance
business in the Current Year, increasing to 2.6% in Year 1 and 2.7% in Year 2, to proxy a mild rise in interest
rates.
Other income
A lot of what banks call other income should be restated below the line as not being very representative of
regular operating business. Earnings from equity-accounted companies (or associates) even if they can be
considered as recurring, should be segregated as they have already been taxed and therefore are not included
in taxable income. The same way, capital gains on the sale of properties should be transferred below the line
as they are not part of the recurring operations of a bank. Once netted from insurance income, associates, and
other non-recurring capital gains, the Other income position should mostly be flat from one year to another.
Figure 18: P&L XYZ revenue block
Year Zero Current Year Year 1 Year 2
Interest income 10.33 10.09 10.22 10.55
Interest expense -6.00 -6.59 -6.45 -6.45
Net interest income
4.33 3.50 3.77 4.10
Net commissions and f ee income 1.50 1.50 1.50 1.58
Trading gains
1.00 1.00 0.80 0.84
Insurance income 0.50 0.44 0.45 0.48
Other income
0.25 0.25 0.25 0.25
NET REVENUES BANK XYZ 7.58 6.69 6.77 7.25

Wrapping up all the forecast work above, we can see that XYZ should post revenues going down 11.7% in the
Current Year, followed by a mild recovery of 1.2% in Year 1 and a more sustainable +7% in Year 2.
The cost block
The cost block is materially easier to assess, as it is composed of two series of costs, divided into three types:
Cash costs:
o Salaries and employee benefits plus other forms of compensation ;
o General & Administrative expenses, also known as non-compensation costs.
Non cash costs:
o Depreciation of tangible and amortization of intangible assets.
The line between these two components may be somewhat blurred to the extent that in some banks (the
investment banks) a part of the employee compensation is being deferred. Despite this, we will consider
salaries and employee benefits as a whole as being cash costs. We note that some banks account for deferred
compensation on the year it is generated, rather than when it is paid.
Total costs = cash costs + non cash costs
For the sake of clarity, goodwill impairment is excluded from the cost base as it can reasonably be considered
as a below the line item since reasonably infrequent.

The principle variables to look at when one forecasts the cost base of a bank are as follows:
Cost income ratio (CIR): sometimes the bank has a CIR target, and therefore the analyst can position the
forecasts as a function of the CIR target, depending how realistic management targets are;
Scope Ratings

February 2014 28 / 65

Compensation costs % total costs and its opposite, non-compensation costs % total costs:
sometimes a bank can give targets with regards to the proportion of administrative expenses versus total
costs. A bank will try to optimize its non-compensation expenses versus the totality of its cost base. A good
analytical indicator is also to monitor the proportion of compensation expenses to revenues. In banks
where the compensation of employees depends on revenues, the sensitivity of costs to revenues will be
higher;
Inflation rate: for retail-based banks, the yearly inflation rate is an important indicator for mandatory
increases in salaries, for example.
Litigation costs: a large component of bank costs post-crisis is the litigation line. Usually, the banks give
guidance on the trends underlying that line, as companies seem to consider these costs as somewhat
exceptional. However, since the crisis and over the last seven years, litigation costs have become more
frequent, and therefore we include them as part of the regular cash non-compensation costs (even if
these costs are most of the time provisions for litigation).
Cost-cutting: in these post-crisis years, a lot of banks have given to the markets cost-cutting targets.
Usually they take the form of multi-year objectives adding to a total of x after a period of, say, five years.
This usually never means (unless it is specifically spelled out by the company) that the cost base in Year 1
will go do down by x in absolute terms in five years time. Usually the bank either (a) increases the salary of
its remaining employees; and/or (b) re-invests parts (or all) of the cost savings into the business. This
means that absolute declines in cost bases are actually pretty rare, except in investment banking where
variable compensation can be slashed at will if revenues are too low. Unsurprisingly, in the latter cases,
decreases in compensation costs are usually lower than decreases in revenues, showing that cost flexibility
at investment banks is only relative.
Figure 19: P&L XYZ cost block static
Personnel expense -4.10 -3.90 -3.79 -3.49
General & Administrative expenses -1.80 -1.70 -1.64 -1.48
Depreciation of tangible assets
-0.30 -0.30 -0.30 -0.30
TOTAL COSTS BANK XYZ -6.20 -5.89 -5.74 -5.27

XYZ has a poor cost control in Year Zero, as its cost-income ratio is close to 82%. The management has
therefore decided to lower the cost base by 15% by Year 2, through a series of branch closures, which should
ease the cost burden of the company. The balance of cost savings will be reached through a control of the non
compensation to income ratio, which the company would like to bring closer to industry standards (20%-25%).
Figure 19 presents the cost base of XYZ on a static basis following the successful completion of the cost
savings program. A 15% cost savings program implies net decrease in costs of 15% x 6.20, so 0.93.
We have assumed that the cost savings program would be implemented by around 33% in the Current Year,
66% in Year 1 and 100% in Year 2. We have also assumed that two thirds of the cost savings program applies
to compensation expenses while the remaining third is taken off the cash non-compensation expense line.
We assume amortization and depreciation unchanged.
Figure 19 gives a useful sanity check about the cost block of XYZ. Considering all the above assumptions, the
total costs of XYZ in Year 2 amount to -5.27 (versus -6.20 in Year Zero). This corresponds to savings of -0.93.
Our forecasts therefore reflect the program quite accurately.
Scope Ratings

February 2014 29 / 65

Figure 20: P&L XYZ cost block dynamic
General & Administrative expenses -1.80 -1.70 -1.67 -1.56
Depreciation of tangible assets
-0.30 -0.30 -0.30 -0.30
TOTAL COSTS BANK XYZ -6.20 -5.89 -5.84 -5.52

Banks frequently use the opportunity of a cost savings program to re-invest in the business. In the case of XYZ,
the revenue situation is way too dire for comfort so we have assumed that all the cost savings were retained in
the Current Year. In Year 1 and Year 2, we assume a cost growth (from the static cost base) of 2% and 5%
respectively, as illustrated in Figure 20.
With this dynamic cost block forecast, we now have our final cost forecasts for XYZ.
Figure 21: Pre-provision P&L Bank XYZ
Interest income 10.33 10.09 10.22 10.55
Net interest income
4.33 3.50 3.77 4.10
Trading gains
1.00 1.00 0.80 0.84
Other income
0.25 0.25 0.25 0.25
General & Administrative expenses
-1.80 -1.70 -1.67 -1.56
Depreciation of tangible assets -0.30 -0.30 -0.30 -0.30
TOTAL COSTS BANK XYZ
-6.20 -5.89 -5.84 -5.52
TOTAL PRE-PROVISION PROFITS 1.38 0.80 0.92 1.73

Obviously, not every bank does run a cost savings program (although a lot of them do currently). In normal
circumstances, an increase in costs reflecting the inflation rate and/or the expected revenue growth plus some
capital expenditures to replace/upgrade some fixed assets is enough.
The cost of risk block
Despite being one of the most important metrics of bank credit analysis, forecasting the cost of risk is not the
most difficult task. As we emphasised when we introduced asset quality in an earlier section, the assessment of
non performing loans and loan loss reserves is based on the situation of a given bank in the credit cycle. The
same is true of the P&L items feeding the loan loss reserves, i.e. the loan loss charges. Depending on the
accounting system used by a given bank, the cost of risk of a bank will include:
Net loan loss charges, i.e. the addition of new estimated provision on a loan or a portfolio of loans minus
the recovery of provisions on loans where the situation has improved;
Charge-offs, i.e. the decision of a bank to consider a debt uncollectible and therefore write-off its amount on
its P&L. When a bank does that, the loan and (if relevant) the associated provision both disappear.
Impairments. An impaired loan is a loan where there has been a deterioration of credit quality which is such
that the bank no longer has assurance as to the timely collection of the full amount of principal and interest.
As we have shown before, goodwill impairment is the result of reflecting in the accounts the fall in the value
Scope Ratings

February 2014 30 / 65

of a past acquisition. In other circumstances, an impairment can impact a loan, but also a security or any
financial instrument where the credit quality of a counterparty has changed due to financial stress. The
provisions that a lot of European banks had to write against specific sovereign debt in 2011 were the result
of the restructuring plan decided upon by European countries and bondholders, and would definitely belong
to that category.
However, some impairments (goodwill or specific sovereign impairment) can be retreated below the line if the
analyst considers them as exceptional/non-recurring enough.
The cost of risk is a function of:
The loan growth;
The temporal position of the bank in the credit cycle, in other words what is the history of the banks cost of
risk ratio (as defined as loan loss charges % Total average gross loans);
And, for some portfolios, the LTV (Loan-To-Value) ratios will be a good indicator of the risk a bank is ready
to take on a borrower.
The loan growth has been forecasted in the balance sheet section.
As for the cost of risk, lets assume that the past five years of Bank XYZ look like this
Figure 22: Past cost of risk of XYZ Bank
("(&%
("&)%
(")(%
("$)%
!"&)%
&%
("((%
(")(%
!"((%
!")(%
&"((%
&")(%
*) *# *' *& *! *+,- 2+-3

Cost-of-risk ratios can be very volatile and their variation can be quite brutal from one bank to another.
There are two ways to forecast the cost of risk. The first one is the easiest and is based on the track record of
the bank at different points in the course of the economic cycle. The second analyses the probability of default
and losses on the banks credit exposure as disclosed in the Pillar III report. Both methods can be used
independently or in association. We favour a combination of the two methods even if we are aware that not
every bank publishes a Pillar III report.
The first method and simpler one, consists in examining the cost of risk history of the bank on a five- to ten-
year view so as to capture the provisioning level over the cycle and, if possible, over several cycles. While doing
this analysis it is possible to identify a mid-cycle cost of risk level, which is the point where the loss trends of a
banks loan portfolio tend to normalize.
Scope Ratings

February 2014 31 / 65

The task will therefore consist in identifying where the bank positions itself at any given point in time vis--vis
this mid-cycle cost of risk. An important point is also to determine the past peak-to-trough costs of risks and to
determine to which extent the current cycle is similar or different to past ones.
In this first way of forecasting cost of risk, the analysts judgment is quite critical to the overall assessment. For
example, and based on Figure 22, it seems that the trough-to-peak variation of Bank XYZs cost of risk (from
virtually zero to 2%) is quite extreme and reflect an important deterioration from a situation that was probably too
benign to begin with. The likelihood of such a credit cycle repeating itself as such will be a function of different
variables, including loan growth, LTVs, the banks credit policy, the macro-economic environment, etc.
The second method and more comprehensive way to forecast the cost-of-risk properly is to include an
analysis of the LGD (Loss-Given Default) and PD (Probability of Default) across different parts of the loan book.
This can only be done for banks reporting Pillar 3 reports among their public disclosure.
LGD is the percentage of loss over a total exposure when the counterparty of a bank defaults. The estimate of a
banks total exposure to a specific counterparty at the time of default of this counterparty is known as EAD
(Exposure at Default). LGD can be calculated easily as Losses % EAD. So typically if we assume that the client
of a bank defaults and that at the moment of default, the EAD was estimated at 1,000,000, and if we also
assume that the bank manages to sell the collateral attached to this transaction for 350,000, the LGD stands at
65%:
(1,000,000-350,000)/1,000,000.
The LGD depends fundamentally on the LTV if the loan is a mortgage. Collateral analysis is very important and
so is covenant analysis for types of corporate loans. Consumer loans tend to be uncollateralized.
PD measures the likelihood of default of a counterparty over a particular time horizon. The PD is a metric that
can be assessed with historic data. Keeping the same example as the counterparty mentioned above, the
default probability of the counterparty is assessed at 25%.
As mentioned above, the best place to source PDs and LGDs is the Banks Pillar 3 report, as the RWA
calculations under Internal Ratings-Based approach (IRB) depend on the Expected Loss (EL) bucket the
company assigns to the various portions of its loan book. The EL is defined as LGD x PD.
In our example above, the EL on our 1,000,000 transaction would be: 65% x 25% = 16.25%.
Since the models used under IRB are internal, they are based on the companys own credit history. It is
therefore a good idea to cross check EL across banks with similar books in similar geographies and check for
inconsistencies or similar ELs.
Finally, some banks are required to publish the difference between their provisions and the ELs as per their IRB
methods. Some companies post severe differences between both metrics, which is a good indicator as to where
the cost of risk may be headed in the future.
In that context, XYZ had virtually zero cost of risk five years ago but this metric deteriorated sharply and after
having reached mid-cycle in the course of Y-3, the asset quality of XYZ deteriorated and its cost of risk soared
to a severe 2% in Year Zero.
Considering the credit cycle history of XYZ and a thorough analysis of the ELs, PDs and LGDs through its Pillar
3 report, and also considering a more benign credit quality environment as well as sustained GDP recovery in
all the countries in which XYZ is operating, XYZs credit quality should start improving. This is also confirmed by
management guidance. What we do not know is the degree of improvement from one year to another. We
model the following:
Scope Ratings

February 2014 32 / 65

Figure 23: Past, present and future cost of risk of XYZ Bank
("(&%
("&)%
(")(%
("$)%
!"&)%
&%
!"')%
("4(%
(")(%
("((%
(")(%
!"((%
!")(%
&"((%
&")(%
*) *# *' *& *! *+,- 2+-3 ./--+01
*+,-
*+,- ! *+,- &

Generally and historically for a bank, a cost of risk of 1% is quite bad (except for consumer lenders, which have a
structurally higher cost of risk). At 2% or above it is catastrophic. So our modelling of a mild decrease until getting
to top-of-mid-cycle number in Y+2 is conservative enough in our view, as demonstrated by Figure 23.
Adding this metric to our XYZ P&L enables us to bring the operational P&L to a close and present a pre-tax
profit (PTP) forecast.
Bar the cost of risk, the only additional line to forecast in the PTP is the Income from associates share of the
results of companies in which the bank has a significant interest. Assuming the perimeter of the group stays the
same from one year to another, this line should be assumed to grow by the estimated profit growth of the
consolidated companies. As one can imagine, this growth can be very different depending on the sector and the
country in which the consolidated company operates. For simplicity reasons, we assume this contribution flat at
XYZ.
Scope Ratings

February 2014 33 / 65

Figure 24: Pre-tax profit breakdown Bank XYZ
Interest income 10.33 10.09 10.22 10.55
Net interest income
4.33 3.50 3.77 4.10
Trading gains
1.00 1.00 0.80 0.84
Other income
0.25 0.25 0.25 0.25
-1.80 -1.70 -1.67 -1.56
-6.20 -5.89 -5.84 -5.52
Cost of risk
-1.95 -1.35 -0.81 -0.52
Income f rom associates 0.25 0.25 0.25 0.25
TOTAL PRE-TAX PROFITS
-0.32 -0.30 0.36 1.46

As we can see, as a result of sharply deteriorating asset quality, XYZ has reported a pre-tax loss in Year Zero.
Despite an improvement in asset quality this Current Year, the bank will in our view continue to be loss-making,
because of weaker revenues. The improvement in profitability should be most visible in Year 2 because of the
impact of the cost-cutting program.
The below-the-line block
This important block is divided in three major components:
The non-recurring items;
The income tax;
Minority interest
The non-recurring items
Under IFRS, there is not really any non recurring below-the-line reporting, while under US GAAP both
exceptional items and extraordinary gains/losses are allowed. There are many non-recurring items to be
identified in a banks P&L and usually analysts are dependent on the banks willingness to identify these items
and communicate on them. They are usually restated from regular revenue and costs lines where they have
to be for accounting purposes, and it is the job of the analyst to restate them below the (pre-tax) line for
analytical purposes.
The most important ones are:
Capital gains (on property or on equity stakes). In that particular case, capital gains on trading or AFS assets
are usually considered as part of the day-to-day operations and are reported in trading gains in the P&L. Non-
recurring capital gains on large properties or on long-term investments are considered non-recurring and
therefore should be restated below-the-line. For example, if XYZ sells its 18% stake in a consumer finance
business, it will appear as revenue, but should be restated below the line. A capital gain on a T-bill held at
market value is definitely a trading gain, and should remain a recurring revenue.
Scope Ratings

February 2014 34 / 65

These non-recurring capital gains are generally impossible to forecast, except when and if they are well-flagged
by the company. They only appear in analysts forecasts for the Current Year in which they have been
announced. So if the sale of that subsidiary was announced in Q1, then it would remain as is for the full year. In
the context of XYZ, the sale of the subsidiary brings a capital gain of 0.6 in the Current Year. There were none
announced in Year Zero we assume none realised in Years 1 and 2.
Restructuring charges. The second most important position among below-the-line items is restructuring
charges. Restructuring charges usually appear as an expense (cost) but they have to be restated from the
regular costs as restructuring charges, and therefore as one-offs. These charges are here to make a cost
savings plan possible, and they represent the costs of this savings program (in particular costs attached to
redundancies etc.). Usually, they represent between 70% and 100% of the achieved cost savings. They can be
accounted for as one-off, but more and more accounting rules demand that the charges be allocated as and
when the cost savings are taking place, so that restructuring charges can take several years to be accounted
for.
To come back on the very specific example of XYZ, we remember that the bank has a cost-savings program of
0.93 to be completed in YEAR 2. We assume for the purpose of this exercise, that to complete the plan XYZ will
have to recognize 0.7 of restructuring charges.
We also assume that these restructuring charges will be recognized proportionately to the implementation time
of the cost savings program, so 33%-66%-100% for Current Year, Year 1 and Year 2 respectively, so it means
charges of roughly 0.23 every annum.
Own debt. Another quantitatively big entry among the non-recurring items is the own debt gains. Even if the
overall economic impact of own debt gains and losses is zero over the lifetime of a bond (as a bond is by
definition repaid at par value at maturity), the variations of the price of the bond in the course of its life can be
significant. Depending on whether the credit spreads of a company widen or tighten, said company will
recognize either gains or losses on its own debt respectively. A gain of 20 is recognized if a bank buys say at 80
a bond which it has issued at 100. The bank has now one of its own bonds on balance sheet for a price of 80. At
this point, if the CDS of the bank narrows sharply, it is likely the price of that bond would increase to say 87.
Versus the former market price of 80, the bank now has to recognize a capital loss of 7. We will note that gains
and losses on own debt are recognized only if this debt is classified in the fair value through profit and loss line
in the balance sheet, so this recognition does not apply to all own debt.
Own debt accounting is volatile and extremely artificial since the end game does not change: the bond matures
at 100 in any case. Also it sends the wrong message: the more credit spreads are widening and the more the
credit quality of a bank deteriorates, the more ample the capital gain the bank can recognize as it is assumed
that the bank can buy its debt in the market more cheaply. As a result of this inconsistency, regulations exclude
own debt gains and losses from Tier 1 recognition. These gains and loses are usually ignored by analysts and
not forecast. But as the year goes by, these items tend to artificially inflate the below-the-line block.
They should therefore be kept well-segregated. For example, we assume that due to the deterioration of its credit
spreads, Bank XYZ generated a capital gain of 0.6 in Year Zero. Considering that the market has remained
cautious about the credit trends of XYZ, the bank has so far recognized 0.4 of own debt gains in the Current Year.
We have decided not to forecast these moves in Year 1 and Year 2 as these gains and losses are deducted
from Tier 1 capital in any case. Therefore, forecasting them seems a bit pointless, even if their relative noise
means that they need to be carefully considered in the course of the analysis of the banks P&L.
It seems IFRS has realised the flaws around own debt accounting. As a result, the International Accounting
Standards Board (IASB) has introduced IFRS 9, which addresses the sensitive topic of hedge accounting.
Regarding own debt, the IASB states that IFRS 9 requires changes in the fair value of an entitys own debt
caused by changes in its own credit quality to be recognized in other comprehensive income rather than in the
P&L. Companies can apply this aspect of IFRS 9 without having to abide by all the other requirements of the
Scope Ratings

February 2014 35 / 65

standards. As a result, it is probable that this line will be less important in the forthcoming years but since
there is no mandatory date for IFRS 9 to be implemented, Scope believes it is more prudent to maintain this line
well identified for the time being.
Credit Valuation Adjustments/Debit Valuation Adjustments/Funding Valuation Adjustments (CVA/DVA/FVA).
CVAs and DVAs are linked to the own debt issues above, but address a slightly different concept. Simply put,
credit valuation adjustment is the amount subtracted from the mark-to-market value of a derivatives position to
account for the expected loss in case of counterparty defaults. For example, if Bank XYZ has contracted a CDS
on a large corporate with Bank ABC, the value of the derivative will have to take into account the probability of
ABC defaulting (and not just the credit fundamentals of the large corporate). In this case, the CVA will be
deducted from the value of the derivative, to reflect the possibility of ABC collapsing.
But a trade is a two-sided story and XYZ will also have to take into account its own creditworthiness to calculate
the value of the derivative properly. In the case of XYZ, the bank, both in Year Zero and in the Current Year, will
have recorded positive DVA gains on the basis of its deteriorating credit position.
As is the case for own debt though, while the bank has to recognize a loss reflecting the credit risk of the
counterparty (CVA), it is also required by the regulator to deduct all DVAs from its Common Equity Tier 1
capital. To this extent, the approach to CVA/DVA is not dissimilar to own debt.
FVA is a related concept. Traditionally, valuation methodologies employed for derivatives reflect the value of
cash-flows discounted by credit risk and funding cost. FVA is a parameter that integrates funding costs in the
valuation of uncollateralized derivatives.
Again, we consider CVAs, DVAs and FVAs as below-the-line items since they reflect very volatile items that
are representative of a perception of a credit risk and only cover the length of a particular trade.
As for own debt, we do not believe it is an item worth forecasting, all the more that it is deducted from the Tier 1
capital calculation.
Goodwill impairment. As the goodwill has to be submitted to impairment tests every annum, effective
impairments are not that frequent, except when market values end up being materially lower than the values at
which the goodwill is booked. Considering its non-recurring aspect, goodwill amortization definitely belongs
below the line.
However, a particular attention should be given to an item called Purchase Price Allocation (or PPA). Following
an acquisition, the excess purchase price paid over the net assets of an acquired company is usually (and
confusingly) called goodwill. In reality, there are two components to this goodwill.
First, the comparison between the identifiable fair value of assets and liabilities purchased and the book value of
these assets. For example, if we assume that a bank has paid 100 for the purchase of a company with net
assets of 20, the acquirer has to identify as much as possible the component of the excess 80. After careful
auditing and valuation of the assets of the purchased company, the company certifies that about 35 can be
attributed to the fair value of property, plant, equipment, securities and identifiable intangible assets (such as
brand/trademark, technology, workforce..). This identified 35 is called Purchase Price Allocation.
The second, unidentified portion of the purchase price in excess of the net assets of the acquired company is
the goodwill proper.
This means that the 100 purchased price of the acquired company can be broken down between: (1) the reported net
assets of the company (20); (2) the PPA or identified intangibles (35); and (3) the goodwill proper (45).
It is very important to identify PPAs because the PPAs are amortisable over the life of the identified assets (no
more than ten years), while the goodwill is not (and is just annually tested for impairment).
Scope Ratings

February 2014 36 / 65

Even if we do not plan to forecast any PPA for Bank XYZ, it is important to know about it as the amortization of
PPA can represent a sizeable non-cash charge to the P&L account. Conversely, if a bank has purchased a
company for less than its net assets, the acquirer will have to recognize a positive item (sometimes very
sizeable) to the P&L, which will artificially boost the results of the acquirer.
Figure 25: "Blow-the-line block" - non recurring items - Bank XYZ
Capital gains 0.0 0.6 0.0 0.0
Restructuring charges 0 -0.24 -0.23 -0.23
Own debt/losses
0.6 0.4 0.0 0.0
CVA/DVA gains/losses 0 0 0 0
Goodwill impairment
0 0 0 0

Taxes and minority interest
Taxes. The regular taxes (VAT, crisis-induced taxes etc.) appear usually at expense level and should remain
there. Near the bottom line, taxes usually refer to consolidated income taxes. Here, the tax rate will usually be a
blend between the different tax rates of the regions in which the bank operates. One has also to take into
account the type of earnings to be taxed (e.g., income from associates is already taxed and therefore not
taxable, Goodwill amortization is not deductible, and capital gains taxes usually have a different rate from
regular income tax).
Banks also operate in many different countries with different tax regimes and are submitted to different DTA
treatments, which mean that forecasting taxes can at times be challenging. However, more complex banks
usually do break down the difference between their normalised tax rates and their real tax rates, making the tax
forecasts somewhat easier.
In the context of XYZ, we assume that, considering its regional earnings mix, XYZ has a weighted-average
income tax rate of 34%. We also believe that the rate is limited to 20% on the recent capital gain.
Lastly, the losses of Year Zero and Year 1 have given rise to tax credits for the company. We assume that the
tax paid on the profits for Year 1 and Year 2 is lessened by the TLCFs of Years 0 and Current Year.
On the basis of these data, we can therefore calculate a real tax rate, which can be materially different from
the blended tax rate as described above. For example, in Year Zero, the operating loss posted by XYZ led way
to no taxes whatsoever. In the Current Year, the taxable profit (before income from associates) was too small
to give rise to any material income tax. Same with Year 1.
In Year 2, we have applied the blended 34% to XYZs profits (excluding income from associates) but we have
also netted the payable amount by the tax credit generated in the course of Year Zero (so about 0.2). As a
result, the real tax rate for Year 2 for XYZ stands at about 15% only.
Minority interests. In theory, the minority interests should follow the trend of the profitability of the company as a
whole; therefore proxying the pre-tax profit growth of the company as a whole to this line makes sense unless
the analyst benefits from divisional minority interest disclosure.
We are now ready to present Bank XYZs full P&L for the Year Zero-Year 2 time period.

Scope Ratings

February 2014 37 / 65

Figure 26: Full Profit & Loss account forecasts of Bank XYZ
Interest income 10.3 10.1 10.2 10.5
Net interest income
4.33 3.50 3.77 4.10
Net commissions and fee income 1.50 1.50 1.50 1.58
Trading gains
1.00 1.00 0.80 0.84
Other income
0.25 0.25 0.25 0.25
-1.80 -1.70 -1.67 -1.56
-6.20 -5.89 -5.84 -5.52
Cost of risk
-1.95 -1.35 -0.81 -0.52
Income f rom associates 0.25 0.25 0.25 0.25
TOTAL PRE-TAX PROFITS
-0.32 -0.30 0.36 1.46
Capital gains 0.00 0.60 0.00 0.00
Restructuring charges
0.00 -0.24 -0.23 -0.23
Own debt/losses 0.60 0.40 0.00 0.00
CVA/DVA gains/losses
0.00 0.00 0.00 0.00
Goodwill impairment 0.00 0.00 0.00 0.00
Income taxes
0.00 0.15 0.00 -0.21
Minority interest -0.05 -0.05 -0.10 -0.25
TOTAL NET ATTRIBUTABLE PROFITS
0.23 0.56 0.03 0.76

This mock P&L of a completely made-up bank is enough to reflect the difficulties of bank analysis in general and
bank forecasting in particular.
On the basis of our forecasts, the underlying fundamentals of XYZ should improve. At the same time though,
the net attributable profit of the company tells a completely different story. A net profit of 0.23 in Year Zero more
than doubles to a profit of 0.56 in the Current Year but falls to virtually zero in Year 1. In reality we know that in
the Current Year the revenue environment should be weaker than in Year Zero and with loan losses remaining
high, it is only through non recurring items (capital gain of 0.6 and own debt gain of 0.4) that XYZ can report a
post-tax profit (while the bank was loss-making at pre-tax level).
It is really at the end of Year 2 that we are expecting XYZ to significantly improve its profitability and report a net
profit at least in line with its improving operating trends (the cost-cutting plan and decreasing provisions in
particular).

Scope Ratings

February 2014 38 / 65

Wrapping-up the forecasts: adjusting the balance sheet for earnings and distribution
After having forecast the P&L in some detail we are now ready to complete the balance sheet forecasts of XYZ.
To do this, we need to increase (or decrease) the capital base of XYZ by the amount of net attributable profit (or
loss) reported for the year, MINUS the dividend distribution for the year, MINUS the share buy backs completed
by the bank in the course of the period under review, PLUS the portion of the dividend that the company will
have decided to pay in scrip.
From the historic position of Year Zero - when the bank decided to pay no dividend in light of its operating
losses, we assume that XYZ will want its shareholders to benefit from the capital gain on the sale of the
consumer credit business in the Current Year. If we assume XYZ will distribute 50% of the pre-tax capital gain
this represents a dividend of 0.3. Out of a net attributable profit of 0.56, this is a payout ratio of 54%.
Despite being just breaking even in Year 1, we will assume that XYZ will still want to reward its shareholders for
their loyalty that year, with another dividend of 0.3 (the payout this time is a multiple of the net profits). With the
strong recovery in Year 2, XYZ should also manage to maintain the payout at an absolute level of 0.3,
representing a payout ratio of 39%.
We note it is important to forecast the expected payout of the company as it enables the credit analyst to
forecast the retained earnings and therefore properly assess the shareholders equity base of a bank arguably
a cornerstone of bank credit analysis.
We adjust the asset side of the balance sheet with the cash position of the bank, to make sure that assets and
liabilities remain perfectly balanced.
Assets Liabilities & shareholders' equity
Government Bonds
85 CDs 95
Equities
80 Current & def erred tax liabilities 3
Current & def erred tax assets 20 Technical provisions 20
Other assets 5 Other provisions 10
2 Other liabilities 5
Intangible assets 2 Total Liabilities 423
Goodwill
7 Shareholders' Equity 20
Fixed assets 2
Total Assets
443 Total Liabilities & Shareholders' Equity 443

Scope Ratings

February 2014 39 / 65

Figure 28: Balance sheet Bank XYZ Current Year final
Government Bonds
72.3 CDs 80.8
Equities
Current & deferred tax assets 16.0 Technical provisions 17.4
Goodwill
Fixed assets 2.0
Total Assets

Figure 29: Balance sheet Bank XYZ Year 1 final
Government Bonds
72.3 CDs 80.8
Equities
Goodwill
Fixed assets 2.0
Total Assets


Scope Ratings

February 2014 40 / 65

Figure 30: Balance sheet Bank XYZ Year 2 final
Government Bonds
75.0 CDs 75.0
Equities
Goodwill
Fixed assets 2.0
Total Assets

We can see that because of a structurally low profitability and a generous dividend distribution, the
shareholders equity of XYZ does not increase much and grows from 20.0 to 20.5 in four years, a limited CAGR
of 0.8%.

Scope Ratings

February 2014 41 / 65

Forecasting the business lines
While the sections above have informed on how to forecast a banks balance sheet and P&L, the degree of
complexity of some business models justifies the use of different forecast techniques.
Scope adopts a business model-based methodology, and for certain types of banks forecasting individual
business lines is the only way to properly assess their future profitability.
At a minimum, banks usually give an abbreviated P&L by business line, i.e.:
Revenues
Costs
Loan loss charges
Pre-tax profits
Some banks provide an individual divisional tax rate. Some allocate directly the taxes and the minority interest
to the unallocated portion of the P&L, also known as the corporate centre.
Some banks segregate their business geographically. Some others segregate by businesses. Some banks give
a minimum breakdown of two very generic business lines while others split their businesses in ten or twelve
different business lines.
Some other even present a matrix-based approach of their financials between geographies and individual
business lines.
Divisional balance sheet 1.0
Generally, the information provided on business lines is sufficient to forecast divisional profitability, but it
becomes way too insufficient at balance sheet level. At a minimum, a bank will provide the capital allocated to
each business line. This, in turn, helps the analyst determining the risk-weighted assets (RWAs) by division.
This assumes that the bank provides the normative common equity Tier 1 on which the allocation is based.
For example, if a bank has two business lines, and we know through public disclosure that the first one is being
allocated a CET1 of 3 and the other one a CET1 of 2, it is easy to calculate the RWAs if the normative CET1
ratio of the bank for the purpose of allocating capital to business lines is 10%. The divisional RWA will be 30
(3/10%) and 20 (2/10%) respectively.
At times, it is nonetheless helpful to have more than the allocated capital as a balance sheet metric for divisional
analysis. Some banks are very good at disclosing divisional balance sheets, while some others are more
reluctant. For the latter, a possible proxy for determining divisional assets lies in analysing the RWAs % Assets
ratio. The analyst should use the RWA % Assets ratio of the last reported period to forecast total assets on the
basis of RWA levels.
Some banks publish their loan book by division. In this case, it can be more practical (and more accurate) to use
the RWA % Loans ratio, also because the bulk of RWAs are credit risks, and therefore tied to the loan book to a
large extent.
To come back to our earlier example we have shown that the two business lines of the bank we used as an
example have estimated RWAs of 30 and 20 respectively. If the ratio of RWAs % Assets stood at 33% on the
last reported quarter, then the total assets of both divisions should be 30/33% and 20/33%, so 91 and 61
respectively.
Sometimes an allocated capital number is the only information available to an analyst to make divisional
balance sheet forecasts and in this case, total assets or total loans are the only balance sheet metrics that
can be assessed with a minimal degree of comfort.
Scope Ratings

February 2014 42 / 65

When and if banks make more information available, then the divisional balance sheet should be forecast via
the same methodology that we used in the balance sheet section of this methodology.
Divisional profitability forecasts
First of all, it is important to identify the different business lines. For convenience purposes, we shall use the
main business lines used by Scope in its methodology:
dRCB (domestic Retail and Commercial Banking)
fRCB (foreign Retail and Commercial Banking)
WIB (Wholesale & Investment Banking)
WAM (Wealth & Asset Management)
We will add also the Corporate Centre/Unallocated items to the forecasts methodology. We could have added
a specialised lending division but to be fair these businesses (mortgage lending, consumer lending, leasing)
are all derivatives of the RCB business, with more wholesale funding and different cost of risks patterns but the
fundamentals and forecasting aspects are broadly similar.
Figure 31 presents the divisional P&L and allocated capital presented by Bank XYZ in Year Zero.
Figure 31: Divisional profitability of Year Zero- Bank XYZ
dRCB fRCB WAM WIB
Unallocated
/corporate
centre
Total
Revenues 2.50 1.10 1.40 2.00 0.58 7.58
Costs -1.95 -0.88 -0.98 -1.40 -0.99 -6.20
Pre-provision profits
0.55 0.22 0.42 0.60 -0.41 1.38
Provisions -1.00 -0.70 0.00 -0.25 0.00 -1.95
Associates
0.00 0.15 0.05 0.00 0.05 0.25
Pre-tax prof its -0.45 -0.33 0.47 0.35 -0.36 -0.32
Other
0.00 0.00 0.00 0.00 0.60 0.60
Taxes 0.00 0.00
Minority interests -0.05 -0.05
Net attributable profits
0.23
Allocated capital
7.00 3.00 1.00 7.00 2.00 20.00
Assets under management 400.00
Cost-income ratio 78% 80% 70% 70% 171% 82%
Return on allocated capital
@normalised tax rate of 35%
(%) -4% -7% 31% 3% -12% -1%

dRCB
If anything, the forecasts attached to the domestic retail banking operations are the ones that are the most
closely replicable to the work we have done for the P&L as a whole in an above section. Usually, the bank will
provide a breakdown of its loan book (mortgage loans, consumer loans, etc.).
There will also be a breakdown of revenues (between net-interest income, non-interest income and sometimes
trading gains but it is likely that the trading in this context reflects pure retail-based execution).
Scope Ratings

February 2014 43 / 65

In the context of Bank XYZ, we have assumed that the deposit war that we alluded to at the beginning of this
report created a 10% revenue fall in dRCB revenues between Year Zero and the Current Year. At the same
time, we have acknowledged the cost savings that the division benefitted from on the basis of the restructuring
plan. Also, we have included in our forecast the proportion of the restructuring charges that we believe are
allocated to this division.
Lastly, a thorough examination of the cost of risk of the division based on the provision history of the division
has enabled us to assess the potential level of loan loss charges in the Current Year, Year 1 and Year 2.
fRCB
Foreign Retail banking again will follow the same analytical pattern as dRCB, with similar disclosure. However,
depending on the markets, the growth rates and the cost of risk will be different. It is frequent in emerging
market economies to post rapid revenue growth rate, but also important cost growth rates, in particular if the
inflation rate in a given economy is high. Costs of risk will tend to be higher and more volatile, while cost-income
ratios will tend to be lower than in mature economies.
As we stand, we have defined XYZ as a bank that was mainly operating in three/four mature/developed
economies. We therefore do not believe that revenue growth expectations should be held up too highly.
Conversely, also because of a high cost-income ratio, we have assumed that the fRCB division would also be
benefitting from the restructuring plan announced at group level. Therefore we are forecasting that about 38% of
the restructuring costs of the plan should be allocated to the fRCB division based on company guidance.
As XYZ applied a similar deposit policy at home and abroad, we have assumed a 4.5% decline in revenues in
the Current Year, followed by roughly a 3% revenue growth in Year 1 and Year 2 respectively. We expect the
cost of risk to experience a steady decline, in line with our analysis at group level.
WAM
Wealth and Asset Management (WAM) covers four different businesses, each one needing a specific forecast
approach:
Asset Management;
Wealth management and private banking;
Custody/Asset administration;
Insurance (life and non-life).
We have already dealt with insurance forecasting in an earlier section.
Asset Management, Wealth Management/Private Banking and Asset Administration are all asset-based
businesses. In other words, for a given amount of assets under management (AuMs), revenues on these assets
under management (also called gross margin) tend to remain quite stable, so between 80bps and 120bps in
Private Banking, and between 30bps and 50bps in Asset Management (depending on the bank). Custody/asset
administration is essentially a volume-driven business and the gross margin levels are very low (about 2bps to
5bps).
For private banking and asset management, the gross margin can be broken down in three components:
A management fee usually the most stable of the revenue components since it is the reflection of a multi-
year mandate;
Transaction fees, reflecting the trades of the client base. Transaction fees are also a function of the type of
mandate undertaken by the client. A brokerage mandate will present more transaction-based revenues
than an advisory mandate, which itself will be more active than a discretionary mandate. As a result, a
brokerage mandate will be low-margin, while a discretionary mandate is a high-margin product. There is a
narrow positive correlation between the workload and the responsibility of the bank on one given mandate,
and the margin.
Scope Ratings

February 2014 44 / 65

Foreign-exchange income is also a very important component of private banking revenues considering that
world-wide clients will transact in a limited number of core currencies.
Historically, private banking margins have proven to be extremely stable with some recent exceptions linked to
the de-risking of private client portfolios post the crisis, and for some very specific countries, the pressure on the
banking secrecy which used to warrant a premium margin.
Nonetheless, the key variable for a reliable forecast of the gross margin remains the total assets under
management (AuMs).
As Figure 32 demonstrates, the assets under management at one given time can be determined as follows:
Beginning of Period (BOP) Assets under management;
Plus/minus Net new money inflows (outflows);
Plus/minus average market performance over the period;
Plus/Minus foreign exchange movements;
Equals EOP (end of period) Assets under management.
The above determination of AuMs is valid for both private banking and asset management. Looking at each item
individually:
The net new money inflows (or outflows) are a function of the franchise of the company, and its strength at one
particular moment in time. Usually, pricing is not a key determinant of money flows, but rather the track record of
the bank over a long period of time, its reputation and the market performance of the different portfolios.
Reputation aspects tend to be more important as a flow factor in private banking than in asset management.
The market performance is a reflection of the outlook that the analyst has on different asset classes (bonds,
money-market funds, equities, alternative assets, property ). Recently, in light of very volatile markets, market
performance has been the main driver of AuMs. In more stable market, usually net new money can become an
important driver of AuM trends. Forecasting market performance for one given portfolio can be made easier if
the bank discloses the asset allocation of its different portfolios. When this is not the case, the forecast has to be
more approximate.
Foreign exchange movements are important, particularly vis--vis the reporting currency of the bank under
review. For a bank reporting in EUR, if the bulk of its AuMs are labelled in USD and if the USD weakens
significantly versus other major currencies, the assets under management of the bank will look weaker in EUR
than they would be in USD.
Figure 32: AuM growth XYZ Year Zero to Year 2
Current Year Y1 Y2
BOY AuMs 400.00 344.81 352.40
Net new Money inf lows (outf low) -60.00 0.00 10.57
"Gross" EOY AuMs
340.00 344.81 362.97
Avg market performance during the year 11.10 7.59 17.88
Avg F/X (weak dollar/EUR Year Zero, Recovery in EUR in Y2)
-6.29 0.00 18.78
"Net" EOY AuMs 344.81 352.40 399.63

For the purpose of this exercise we will assume that XYZ has no insurance business and that its WAM business
is a combination of Asset Management and Private Banking. To take the example of Bank XYZ and starting
from a position of total AuMs of 400 in Year Zero. In the Current Year, XYZ has ben pretty vocal about
significant client outflows following the bad performance of a discretionary portfolio. The bank made clear that its
WAM division would lose 15% of its assets in the course of the year. At the same time, capital markets
Scope Ratings

February 2014 45 / 65

improved a little bit and we believe that XYZ will record a performance of 3% in the course of the Current Year
(amalgamating the estimated weighted-average asset class performance of all the portfolios). Lastly, let us
assume that economists are forecasting a marginal decrease of the value of the USD versus XYZs local
reporting currency and, as it happens, a lot of XYZs assets are labelled in US dollar.
Figure 33: Assumed gross margin XYZ
("')%
("'&%
("'#%
("''%
("'!%
("'&%
("'&%
("''%
("''%
("'#%
("'#%
("')%
("')%
("'5%
*+,- 2+-3 ./--+01 *+,- *+,- ! *+,- &

In Year 1 (and following company guidance), we believe that the outflows will have stopped but that XYZ will not
be able to collect new clients. At the same time, market performance should continue steadily (at +2%) while we
do not expect severe F/X changes. Altogether, XYZs AuMs should be up 2.2% versus the Current Year, to 352.
For Year 2, it is assumed that net new money should start improving with a collection of 11 (so about +3%)
while market performance should accelerate (+5%). Part of this gain is compound by a positive 5.25% forex
effect, due to an expected strengthening of the USD against XYZs reporting currency. AuMs are therefore up
by a healthy 13% in Year 2 to 399.6 almost back to Year Zero level despite the loss of the mandate. This
example shows that forex movements can have a distorting impact on the performance of an asset
management division.
Once the work on the AuMs is done, there is not much to do on the revenue side: the margin on AuMs stood at
35bps in Year Zero. We assume that with the loss of high-margin portfolios, the gross margin level on average
AuMs will have declined by about 3bps in the Current Year, before stabilising at around 33-34bps in Year 1 and
Year 2. As one can see, we do not expect strong margin volatility for XYZ. The lower margin in Year 2 is due to
the fact that the revenues in Year 2 do not fully capture the net new money growth in Year 2 yet, nor the inflated
F/X impact.
As for the costs, it all depends on the model under which the bank operates. In private banking, brokerage-
based models (i.e. models where the bulk of the revenues are generated by client transactions) tend to post
higher cost-income ratios than advisory-based models (where the bulk of the revenues are made of
management fees). Indeed, brokers tend to be compensated on the basis of generated income, which triggers
higher cost bases.
After an acceptable cost-income ratio of 70% in Year Zero, we would assume that in the Current Year, XYZ
will not be interested in adjusting its cost base yet on the basis of its willingness to hang on to its remaining
Scope Ratings

February 2014 46 / 65

market shares. But following another moderately successful year in Year 1, we would expect the bank to start
cutting costs in Year 2.
WIB
With WAM, WIB (Wholesale & Investment Banking) is the only division where the forecasts rules are slightly
different from the regular bank forecasts that we have already expanded on. Effectively, WIB can be broken
down in two different sub-business lines:
Corporate banking
Advisory & Capital Markets
Corporate Banking is arguably a banking business which is very similar in methodology to the Retail Banking
business. In both cases, the businesses are quite balance-sheet intensive (since they both include loans), they
both are in part funded with deposits (corporate deposits are more volatile than retail deposits), and there are
also specific corporate costs of risks, depending on specific business cycles. The revenue structure will also not
be dissimilar, with net interest income and fees representing the bulk of the income in this division.
Advisory & Capital Markets is a completely different activity mix as it gathers businesses that are either less
capital-intensive (advisory does not necessitate any allocated capital) or very demanding in regulatory capital
(like Fixed Income, Currency & Commodities (FICC) businesses).
Advisory and Capital Markets can be further divided between:
Primary business
o Advisory/M&A;
o Equity capital markets (ECM);
o Debt capital markets (DCM);

Secondary business
o Equities trading;
Cash Equity
Equity Derivatives
Prime Brokerage
o Fixed-Income, Currency & Commodities (FICC) trading
Rates;
Credit;
Structured Products;
Currency
Commodities
The ranking of the five Advisory & Capital Markets divisions as spelled out above can be read from least
capital intensive to most capital intensive.
While Advisory/M&A is just what it says, ECM and DCM are more capital intensive as the bank can commit to
underwrite some of the issues it manages until they are sold to the marketplace.
Equities trading is not a very capital intensive business, except for prime brokerage, where the bank can be in a
position to offer financing.
Overall, the vast majority of the allocated capital of the Advisory and capital markets sub-division can safely be
said to be allocated to FICC. This assumption tends to make the forecasting work a bit easier.
Scope Ratings

February 2014 47 / 65

Indeed, it is usually impossible to get to a level of granularity that would enable the analyst to analyse in detail
every sub-division of the Advisory and Capital Markets division. Therefore the forecast work is usually limited
to three large revenue blocks:
Investment Banking (combination of Advisory, DCM and ECM);
Equities trading;
Fixed-Income trading.
So overall, Bank XYZs WIB division as a whole would look like this:
Figure 34: WIB breakdown Bank XYZ Year Zero
Corporate Bank
Advisory & Capital
Markets
TOTAL
Corporate Bank Revenues 0.60 - 0.60
Investment Banking Revenues - 0.20 0.20
Equities Trading Revenues
- 0.40 0.40
FICC Trading Revenues - 0.80 0.80
Revenues
0.60 1.40 2.00
Costs -0.30 -1.10 -1.40
0.30 0.30 0.60
Provisions -0.25 0.00 -0.25
Pre-tax profits
0.05 0.30 0.35
Allocated capital
2.50 4.50 7.00

The question is then how do we forecast the WIB division of XYZ:
For the corporate bank, the reasoning is not materially different from the dRCB and fRCB methods that we
have alluded to above. The credit cycle is different, and the cost structures tend to be lower (with cost-income
ratios of 40% to 60% over the cycle simply because a corporate lending division within an investment bank
does not have to feed a large branch network).
The Advisory & Capital Markets division requires a different approach.
In revenue terms, the forecasting work will often be more art than science.
In equities trading and investment banking in particular, the short-term trends can be identified by looking at
data providers disclosing series of league tables, volumes, market shares, etc. But these will essentially be
helpful in predicting advisory and primary capital markets revenues (ECM and DCM).
Equity trading revenues are essentially more difficult to forecast accurately, and will eventually derive from
economic growth pointers, but since markets pay great attention to forward-looking indicators it is probable that
strong equity volumes will anticipate a future GDP growth rate that in itself will drive future corporate earnings.
The reality of the cash equity market though is that its margins are persistently dwindling due to increased
competition and much better trading technology (Direct Market Access - DMA) and client protection rules
(Markets in Financial Instruments Directives - MIFID). Equity derivatives and prime brokerage do, to some
extent, offset this phenomenon since they are higher-margin businesses but the growing importance of CCPs
for the clearing of OTC derivatives cannot be good news for future margins of equity derivatives.
Therefore Equity revenue margins should in theory decline but as this business is an intensely volume-driven
business, revenues are likely to be very cyclical. .
Scope Ratings

February 2014 48 / 65

Fixed Income Trading revenues are even more difficult to capture than equities because the underlying asset
classes are very different: bonds, interest rates-swaps, CDSs, currency, commodities. Fixed income instruments
are also, by their nature, very fragmented, which means that the idea of one integrated global fixed income
market (like the cash equity market for example) is an impossibility. Therefore, a lot of the transactions will be
done OTC, implying that margins are higher. Capital costs are also higher as banks maintain a lot of fixed-
income instruments in inventory as the liquidity of each instrument is limited.
This added complexity is also an advantage, as the capital intensity means that returns can be calculated. If
returns can be calculated, they can be compared over different past cycles, and therefore they can be replicated
in the future even if there is a fair share of volatility in this market.
If we come back to some of the assumptions we made on the divisional balance sheet, we assume that the
allocated capital of the Advisory & Capital Markets division is benchmarked against a core Equity Tier 1 of 10%.
Therefore for the A&CM division, the corresponding RWAs amount to 45 in Year Zero (4.5/10%). If, as we also
said in this section, the RWA % Assets stand at 33%, then we can proxy that the total assets of the A&CM
division stand at 136 (45/33%).
If we now assume that the totality of this capital is allocated to FICC trading (not a bad assumption considering
the capital intensity of that business), then we can deliver an estimated FICC return. At Year Zero, for Bank
XYZ, the gross income return on assets will be: 0.80/136 = 0.59%.
While the gross margin on a Private Banking or an Asset Management business can be compared throughout
the industry (as the reporting and the accounting are reasonably homogeneous), it would be quite dangerous to
do the same with the FICC gross margin, as the internal models that are used to determine the risk-weights of
FICC assets are vastly different from one bank to the other. However, such a number will be helpful to
determine trends within one bank (assuming that the bank has not too materially changed its internal modelling
from one year to another - and this can be seen through the RWA % Assets ratio).
We have imagined Bank XYZs historical FICC margin over the last five years (finishing with Year Zero). Maybe
this trend will help us figure out what is coming next for XYZs FICC margin.
Figure 35: FICC margin Bank XYZ, last five years
("((%
(")(%
!")(%
!"6)%
!"((%
(")6%
("((%
(")(%
!"((%
!")(%
&"((%
&")(%
*) *# *' *& *! *+,- 2+-3

As we can see in Figure 35, the FICC margin of XYZ has shown considerable volatility and this type of
volatility is extremely difficult to duplicate when one makes FICC forecasts. To be fair, a very well-guided
monetary environment where any suggestion of interest rates going up is carefully prepared and communicated
to the widest possible audience, should reduce FICC volatility. At the same time, the simple idea of tapering
Scope Ratings

February 2014 49 / 65

created some notable volatility on FICC businesses world-wide last summer showing that managing the
impact of monetary policy on banks profit and loss accounts is as much an art as it is a science.
On the cost side, an important aspect of the Advisory & Capital Markets business is the level of
compensation. For some banks where WIB operations are important, personnel expenses are disaggregated.
Issues around non-cash based compensation and deferred compensation have to be addressed and the ratio of
compensation expenses % income can be analysed in more details.
For the purpose of this report though, we will stick to the cost-income ratio at divisional level.
Obviously, there are (theoretically) no loan loss provisions for investment banks but often securities impairments
will be recorded on this line.
We are now ready to draw forecasts on Bank XYZs WIB division.
Figure 36: WIB breakdown Bank XYZ Current Year
Corporate Bank
Advisory & Capital
Markets
TOTAL
- 0.40 0.40
Revenues
0.60 1.40 2.00
Costs -0.30 -1.10 -1.40
0.30 0.30 0.60
Provisions -0.15 0.00 -0.15
Pre-tax profits
0.15 0.30 0.45
Allocated capital
2.50 4.80 7.30

Figure 37: WIB breakdown Bank XYZ year one
Corporate Bank
Advisory & Capital
Markets
TOTAL
- 0.34 0.34
Revenues
0.62 1.14 1.76
Costs -0.34 -1.19 -1.53
0.28 -0.05 0.23
Provisions 0.09 0.00 0.09
Pre-tax profits
0.37 -0.05 0.32
Allocated capital
2.50 4.50 7.00

Scope Ratings

February 2014 50 / 65

Figure 38: WIB breakdown Bank XYZ year two
Corporate Bank
Advisory & Capital
Markets
TOTAL
- 0.60 0.60
Revenues
0.65 1.48 2.13
Costs -0.36 -1.09 -1.45
0.29 0.39 0.68
Provisions -0.02 0.00 -0.02
Pre-tax profits
0.27 0.39 0.66
Allocated capital
2.50 4.50 7.00

Corporate bank forecasts: The Current Year shows no revenue progression versus Year Zero considering that
XYZ has decided to reduce its loan growth following difficult years, as we have explained in the first part of this
report. This restrictive policy has been applied to all the divisions of the banks, not only dRCB and fRCB.
Following the Current Year, Year 1 and Year 2 show revenue growth that is in line with the growth of the loan
book (+3.5% to +4% respectively). We have kept the cost income ratio unchanged at 50% in the course of the
Current Year, but have considered that increased market shares in Year 1 and Year 2 would be funded by a
faster expense growth, hence the cost income ratio growing to 55% in these two years. Lastly, in light of a
strongly improving credit situation on the corporate side, we have lowered the cost of risk of XYZ from a high
level in Year Zero to virtually zero in Year 1 and Year 2 (including even net releases in Year 1).
Advisory & Capital Markets forecasts: Management made clear that in the Current Year the Advisory &
Capital Markets revenues would be in line with Year Zero. Considering that the allocated capital to the division
is slightly higher in the Current Year than in Year Zero we choose to slightly decrease the gross FICC margin of
XYZ to offset the impact of the higher capital and therefore maintain FICC revenues stable.
In Year 1 it seems inevitable that interest rates will not be held at their current levels. Therefore forecasting a
reduction in FICC margins of around 20% seems like a reasonable thing to do: this number is consensual
enough to lead to moderate upgrades or downgrades if the situation is materially better/worse than expected.
We expect that higher interest rates are likely to have a negative impact on equities and investment banking
deals as well, leading to a reduction in equity revenues and investment banking revenues by 10% to 15%. At
the same time, as WIB was not part of the restructuring plan that XYZ announced at the end of Year Zero, the
bank was not able to reduce its cost base versus the previous year, which means that the Investment Banking
& Capital markets division ended up being loss-making at pre-tax level in Year 1.
In Year 2, the main beneficiary of a better economic environment validated by higher interest rates has to be, in
our view, the equity markets. This is why we are expecting the revenues of this business to surge by 76%. We
kept the pure investment banking revenues flat, as we expect that the revenue recognition of XYZs primary
deals will probably lag. Lastly, on the back of good markets, fixed income should benefit too but to a lower
extent we therefore choose to moderately increase the FICC margin from the low base of Year 1 but on the
basis of Figure 39, we can see that the analyst expects XYZs FICC margins to remain at historically low levels.
Maybe these low levels represent a new normal for FICC profitability in a Basel 3 world.
Scope Ratings

February 2014 51 / 65

Figure 39: FICC margin Y-5 to YEAR 2 (%)
("((%
(")(%
!")(%
!"6)%
!"((%
(")6%
("))%
("#)%
(")!%
("((%
(")(%
!"((%
!")(%
&"((%
&")(%
*) *# *' *& *! *+,- 2+-3 ./--+01
*+,-
*+,- ! *+,- &

Corporate centre
We are now getting to the black box of bank accounting: the corporate centre. The corporate centre (CC)
gathers the unallocated costs, headquarters costs, central functions (Investor Relations, CFO ) but also group
treasury. Very often the CC also perceives dividends of equity stakes, and can also consolidate non-core
businesses (legacy portfolios, private equity, property companies etc.). Sometimes these non core businesses
are disclosed, sometimes they are not.
Depending on the degree of disclosure pertaining to the CC, the management then has some leeway to adjust the
banks divisional reporting. This has happened a lot in Europe in recent years and a lot of banks have proceeded
to restatements, sometimes on a very frequent basis. In some cases, some treasury functions have been directly
allocated to the divisions leading to a deterioration of the corporate centres results. Indeed, the vocation of the
CC is to post negative pre-tax profits as the CC carries the cost of doing business at group level.
The black box syndrome is all the more potent that every division will have to negotiate transfer costs with the
corporate centre. For example, in some cases, group general counsel will have to help the dRCB division. It is
important for the division to pay the CC for its services at arms length so that the divisions are comparable
between different banking groups. Unfortunately we have no guarantee that these transfer prices remain stable
in time within each bank, the same way we do not know whether the perimeter of the CC is consistent from one
period to the other.
Also, the CC is frequently the centrepiece for all the one-off items happening in the P&L in one given time period
(DVA, own debt, capital gains). As a result, it is necessary to restate a lot of the CCs one-off items below the
line to be able to have a less muddy view about this division.
Until recently, on an annual basis, and restated from below-the-line items, CC results were reasonably
consistent. Clearly there were some cyclicalities (the coupon paid on a group subordinated debt or some
dividends perceived would not recur at the same time in the course of the year) but annually there was some
consistency.
This consistency has to a large extent remained on the expense side but since the cost of liquidity for the banks
has increased in particular the fact of having a lot of money held at Central Banks for a very low return, CC
revenues have in turn become more volatile.
Scope Ratings

February 2014 52 / 65

While until recently it was common practice to proxy the last reported years corporate centre P&L to the
forthcoming years, now it is more prudent and frequent to use averages. In the case of XYZ, the increased
liquidity cost has driven a collapse in revenues from Year Zero to the Current Year. We believe that liquidity
commitments should be slightly less stringent going forward which is why our CC revenues reflect an average
between the revenues of Year Zero and the revenues of the Current Year. On the cost side, we have left the
number unchanged, as it was consistent in Year Zero and in Year 1.
Putting it all together: art and science again
Finally, the business line forecasts of XYZ can be put together: they should look like this:
Figure 40: Divisional profitability of Current Year Bank XYZ
dRCB fRCB WAM WIB
Unallocated
/corporate
Total
Revenues 2.25 1.05 1.20 2.00 0.19 6.69
Costs -1.70 -0.78 -0.98 -1.40 -1.03 -5.89
0.55 0.27 0.22 0.60 -0.84 0.80
Provisions -0.70 -0.50 0.00 -0.15 0.00 -1.35
Associates
0.00 0.15 0.05 0.00 0.05 0.25
Pre-tax profits -0.15 -0.08 0.27 0.45 -0.79 -0.30
Other
-0.15 -0.09 0.00 0.00 1.00 0.76
Taxes 0.15 0.15
Minority interests
-0.05 -0.05
Net attributable profits 0.56
Allocated capital 7.00 3.00 1.00 7.30 2.00 20.30
Assets under management 357.00
Cost-income ratio
76% 74% 82% 70% 542% 88%
(%)
-1% -2% 18% 4% -26% -1%

Source: Scope Rating
Scope Ratings

February 2014 53 / 65

Figure 41: Divisional profitability of Year 1 Bank XYZ
dRCB fRCB WAM WIB
Unallocated
/corporate
Total
Revenues 2.36 1.08 1.18 1.76 0.39 6.77
Costs -1.60 -0.70 -0.98 -1.53 -1.03 -5.84
0.76 0.38 0.20 0.23 -0.65 0.93
Provisions -0.50 -0.40 0.00 0.09 0.00 -0.81
Associates
0.00 0.15 0.05 0.00 0.05 0.25
Pre-tax profits 0.26 0.13 0.25 0.32 -0.60 0.37
Other
-0.15 -0.08 0.00 0.00 0.00 -0.23
Taxes 0.00 0.00
Minority interests
-0.10 -0.10
Assets under management
365.00
(%) 2% 3% 16% 3% -19% 1%

Figure 42: Divisional profitability of Year 2 Bank XYZ
dRCB fRCB WAM WIB
Unallocated
/corporate
Total
Revenues 2.41 1.11 1.22 2.13 0.39 7.25
Costs -1.50 -0.65 -0.89 -1.45 -1.04 -5.53
0.91 0.46 0.33 0.68 -0.66 1.72
Provisions -0.30 -0.20 0.00 -0.02 0.00 -0.52
Associates
0.00 0.15 0.05 0.00 0.05 0.25
Pre-tax profits 0.61 0.41 0.38 0.66 -0.61 1.45
Other
-0.15 -0.08 0.00 0.00 0.00 -0.23
Taxes -0.21 -0.21
Minority interests
-0.25 -0.25
Assets under management
375.00
(%) 6% 9% 25% 6% -20% 5%

Clearly the most important aspect is that the divisional forecasts must match the group balance sheet and P&L
formats. In other words, the sum of all divisional forecasts must equal the group consolidated balance sheet and
P&L.
Scope Ratings

February 2014 54 / 65

More specifically, the group P&L and balance sheets forecasts (for the banks that are not mono-business or
mono-geography) must be driven by the divisional numbers.
Often divisional reporting is less detailed than consolidated reporting. When this is the case, it can happen that
for forecasts purposes the revenues will not be disaggregated between net interest income, fee income and
trading income. Also, personnel and non personnel expenses are sometimes not split by division. Same thing
with the below-the-line items.
It is therefore the analysts job to estimate the accounting breakdown of divisional forecasts, so that they all add
up to the same number as we can demonstrate here: the divisional P&L numbers of Figures 40, 41 and 42
add up to the consolidated P&L forecasts of Figure 26.

Regulatory forecasts
There is one last layer of forecasting left to be done in the context of a proper credit ratings job: forecasting
regulatory ratios. The calculation can be severely complicated, all the more that banks disclosure is often not up
to expectations.
These ratios are to a large extent templated by regulators so that for the analyst, it is just a matter of filling the
form properly and check the results versus the companys own forecast or targets.
Since the start of the crisis, regulators throughout the world have relentlessly worked on two major issues,
widely known to have been at the core of the banking sectors problems since June 2007: capital and liquidity.
Our regulatory forecast work will therefore focus on the four enforced regulatory ratios, two for capital
assessment, and two for liquidity assessment knowing that for some of these ratios (the two liquidity ratios in
particular), current disclosure makes it very difficult to assess the level of these ratios. For the moment, an
analyst using public disclosure can only forecast components of these ratios, and not the full ratios themselves.
The last section of this forecast methodology will include:
Calculation of Basel 3-compliant risk-weighted capital ratios (according to CRD 4/CRR and using the
post 01.01.2018 disclosure of the rules text Composition of capital disclosure requirement published in
June 2012);
Calculation of the Basel 3-compliant unweighted Leverage Ratio;
Net Stable Funding Funding Ratio (NSFR);
Liquidity Coverage Ratio (LCR) or at least some of its most easily identifiable components.
Basel 3-compliant risk-weighted capital ratios
In Figure 43, we are attempting to replicate (and simplify) the template provided by the Basel Committee on Banking
Supervision (BCBS). We assume a post January 2018 template (so basically a fully-phased Tier 1 ratio).

Scope Ratings

February 2014 55 / 65

Figure 43: Basel 3 capital calculation (post 1 January 2018 disclosure template)
Figure 43: Post 1 January 2018 Disclosure template

Common shares issued plus related stock surplus
+Retained earnings
+Accumulated OCI (and other reserves)
+Restated Minorities (Basel 3 definition)
=Common Equity Tier 1 capital before regulatory adjustments (1)

Prudential Valuation adjustments
+Goodwill and other intangibles
+DTAs based on future profitability (TLCFs)
+Cash Flow Hedge reserves
+/-AFS reserves (add-back losses , deduct gains)
+Shortfall of provisions to Expected Losses (EL)s
+Securitisation gain on sale
+Gains and losses due to changes in own credit
+Defined-benefit pension fund net assets
+Own shares (if not netted in CET 1 capital defined above)
+Cross-holdings
= "straight" regulatory adjustments (2)

Common Equity Tier 1 before threshold deductions (3) = (1) - (2)

Investments in financial companies (banks, insurance...) where bank stake is <10%
+Investments in financial companies (banks, insurance...) where bank stake is >10%
+Mortgage servicing rights
+DTAs arising from temporary differences
= "Specified items" as per Basel 3 Annex 2, p. 65 (4)

Maximum combined recognition of specified items (5) = 17.65% x ((3)-(4))

Investments in financial companies (banks, insurance...) where bank stake is <10%- capped at 10% of (3)
+Investments in financial companies (banks, insurance...) where bank stake is >10% - capped at 10% of (3)
+Mortgage servicing rights - capped at 10% of (3)
+DTAs arising from temporary differences - capped at 10% of (3)
= Maximum Recognition of "Specified items" (6)

AMOUNT OF SPECIFIED ITEMS RECOGNIZED IN CET1: (7) = LOWER OF (5) OR (6)

ADJUSTMENT TO CORE EQUITY TIER 1: (8) = (7) - (4)

COMMON EQUITY TIER 1 (9) = (3) + (8)

Additional Tier 1 Capital (10)

TIER 1 CAPITAL (11) = (9) + (10)

TIER 2 CAPITAL (12)

TOTAL CAPITAL (13) = (11) + (12)

Risk Weighted Assets (14)
Source: BIS, Scope Ratings
Scope Ratings

February 2014 56 / 65

This table is sourced from a combination of data and definitions extracted from CRD 4 and the Basel 3
regulatory frameworks, complemented by our attempts at simplifying some lines. The main analytical
adjustments to be made are:
The AFS reserve within the OCI. These are to be excluded from CET1 calculation, so that the gains are
deducted and the losses are added back. These adjustments are in any case phased out by CRD 4 until
31.12.2017
Minority interests: these are eligible to be recognized as CET1, minus the amount of the surplus
common equity Tier 1 of the subsidiary attributable to the minority shareholders (Basel 3, article 62). The
way to take this aspect into account is usually to consider that 50% of the minority interests of a bank are
eligible for CET1. This is a convention generally accepted by the banks.
Derogations: Pretty much every line under CRD 4 is susceptible to an exemption or a delay. These
exemptions and delays need to be taken into account because they can be significant and sometimes they
prevent banks from being comparable on a cross-border basis. A well-known example is insurance, which
in theory should be deducted under the threshold method spelled out on Figure 43. However, Article 471 of
CRD 4 exempts banks from such deductions until YE 2022. Instead, these equity holdings can be
alternatively risk-weighted at 370%.
Risk-Weighted Assets: Basel 3 RWAs can only be forecast with the help of company guidance
(particularly on the additional weight of counterparty risk). To be able to forecast RWAs properly, the
analyst will have to be very mindful of the achievability of RWA reductions, as well as the way some
portfolios are weighted under Pillar 3. A good top-down way to assess the degree of conservatism in a
banks RWA accounting is to assess the evolution of the RWAs % Assets ratio over time.

Basel 3-compliant unweighted Leverage Ratio
The leverage ratio has been the object of a very strong debate over the last few months. The final rules of the
BCBS on the Basel III leverage ratio have clarified the composition as well as the absolute level of the leverage
ratio.
In its simplest exponent, the Leverage Ratio (LR) can be defined as:
LR = Capital Measure / Exposure Measure.
Basel 3 has defined the minimum ratio to stand at 3% until January 1, 2017.
The capital measure is the Tier 1 capital as it is calculated in Figure 43 (item #11).
The exposure measure is a combination of (a) on-balance sheet items; (b) derivatives; (c) securities financing
transactions (SFTs); and (d) off-balance sheet (OBS) items.
Banks must include all on-balance sheet items in their exposure measure including on-balance sheet
derivatives collateral and collateral for SFTs, with exceptions noted below. On-balance sheet items deducted
from Tier 1 are also deducted from the exposure measure to show consistency with capital ratios calculations.
Derivatives create two types of exposure: (1) the exposure arising from the underlying of the contract and (2)
the counterparty credit risk exposure.
The Positive Replacement Value of the derivative (reported on the asset side of the balance sheet) can be used
as a proxy for the replacement cost of the contract. The problem is that banks need to include an add-on for
potential future exposure over the remaining life of the contract. This add-on is a percentage which is applied to
the notional principal amount of the derivative which means an off-balance sheet factor is included. Versus the
consultative document published by the BCBS in June 2013, the January 2014 final framework authorises some
Scope Ratings

February 2014 57 / 65

derivatives exposure reduction, particularly the cash portion of variation margin and trades cleared through
central counterparties both in specific circumstances.
SFTs are included in the exposure measure, but some cash payables and cash receivables in SFTs with the
same counterparty may be measured net if specific criteria are met. Qualifying master netting agreements may
be used in certain circumstances to reduce the exposure.
The off-balance sheet items include commitments (including liquidity facilities), whether or not unconditionally
cancellable, direct credit substitutes, acceptances, standby letters of credit and trade letters of credit. Depending
on the risk profile of these items, a credit conversion factor will be applied to the nominal value of these items.
Some of the elements composing the exposure measure will be difficult to estimate, in particular the derivatives
add-ons. However, considering the fact that banks are starting to report their fully-phased CRD 4 leverage
ratios, and considering that the Tier 1 capital is reasonably easy to identify, the exposure measure can be
deducted and used to forecast leverage ratios going forward, on the basis of the balance sheet growth expected
for the company. Also, considering that the numerator can be easily estimated in the situation where the
company discloses the leverage ratio, the exposure measure can be assessed with reasonable confidence
and hence, by difference, the LR-compliant derivatives exposure. This number can then be compared with the
on-balance sheet present value of derivatives to ease the forecasts going forward.
Clearly, the reading of the Basel 3 leverage ratio is different from one country to another, and Scope ratings will
make sure to properly identify the national differences when forecasting the leverage ratio.
Net Stable Funding Ratio (NSFR)
The Net Stable Funding Ratio (or NSFR) is one of the two liquidity ratios introduced by Basel 3 (with the LCR). It
is the easier of the two to calculate and estimate, but also the most controversial, as it attempts to draw
limitations to the core business of the bank: transformation. This is why, versus the more complex LCR, the
NSFR is still under observation period by the BCBS and remains subject to review to address any unintended
consequences. Despite this delay, the BCBS still intends for the NSFR to become a minimum standard by
January 1, 2018.
Our forecast work takes into account the BCBSs most recent paper on the subject, which is the consultative
document issued by the Committee in January 2014.
As per the Basel 3 International Framework for liquidity risk measurement, standards and monitoring the
NSFR is defined as:
Available amount of stable funding/Required amount of stable funding 100%
The NSFR includes required amounts of stable funding for all illiquid assets and securities held, regardless of
accounting treatment. Additional stable funding resources are also required to support at least a small portion of
the potential call on liquidity arising from off-balance sheet commitments and contingencies.
As mentioned by the liquidity framework document, the objective of the standard is to ensure stable funding on
an on-going, viable entity basis over one year in an extended firm-specific stress scenario.
Computing the numerator: the available amount of stable funding
The available amount of stable funding is calculated by first assigning the carrying value of an institutions equity
and liabilities to one of five categories as presented in Figure 44 below. The amount assigned to each category
is to be multiplied by an ASF (Available Stable Funding) factor and the total ASF is the sum of the weighted
amounts.
Basel defines Stable deposits as deposits that are fully covered by an effective deposit insurance scheme or
by a public guarantee that provides equivalent protection and where the depositors have established
Scope Ratings

February 2014 58 / 65

relationships with the bank that make deposit withdrawal unlikely; or the deposits are in transactional accounts
(e.g. accounts where salaries are automatically deposited).
Less Stable deposits are defined as deposits that are not covered by an effective deposit insurance scheme or
sovereign deposit guarantee; high-value deposits; deposits from sophisticated or high net worth individuals;
deposits that can be withdrawn quickly; and foreign currency deposits.
Interestingly, versus the former proposal, Central Bank funding can be included with an ASF factor of 50%
(versus 0% previously), while the ASF factor for stable and non stable deposits is higher.
Figure 44: Components of Available Stable Funding and Associated ASF Factors
ASF
Factor
Components of ASF category
100%
Total regulatory capital
Other capital instruments and liabilities with effective residual maturity of one year or more
95%
Stable non-maturity (demand) deposits and term deposits with residual maturity of less than
one year provided by retail and SME customers
90%
Less stable non maturity deposits and term deposits with residual maturity of less than one
year provided by retail and SME customers
50%
Funding with residual maturity of less than one year provided by non financial corporate
customers
Operational deposits
Funding with residual maturity of less than one year from sovereigns, public sector entities
(PSEs), and multilateral and national development banks
Other funding with residual maturity of not less than six months and less than one year not
included in the above categories, including funding provided by central banks and financial
institutions
0%
All other liabilities and equity not included above categories, including liabilities without a stated
maturity
Derivatives payable net of derivatives receivable if payables are greater than receivables
Source: BIS
Computing the denominator: the required stable funding for assets and on-balance sheet exposure
The amount of required stable funding is measured based on the broad characteristics of the liquidity risk profile
of an institutions assets and OBS exposure. The required amount of stable funding is calculated by first
assigning the carrying value of an institutions assets to the categories listed on Figures 45a and 45b. The
amount assigned to each category is then multiplied by its associated Required Stable Funding (RFS) factor
and the total RFS is the sum of the weighted amounts added to the amount of OBS activity (or potential liquidity
exposure) multiplied by its associated RFS factor.
The definitions now used by the NSFR are totally consistent with the concepts of High Quality Liquid Assets
(HQLAs) defined for the determination of the Liquidity Coverage Ratio and addressed in detail in the next
section.

Scope Ratings

February 2014 59 / 65

Figure 45a: Composition of Asset Categories and associated RFS factors.
RSF
Factor
Components of RSF category
0%
Coins and banknotes
All central bank reserves
Unencumbered loans to banks subject to prudential supervisions with residual
maturities of less than six months
5% Unencumbered Level 1 assets, excluding coins, banknotes and central bank reserves
15% Unencumbered Level 2A assets
50%
Unencumbered Level 2B assets
HQLA unencumbered for a period of six months or more and less than one year
Deposits held at other financial institutions for operational purposes
All other assets not included in the above category with residual maturity of less than one
year, including loans to non-bank financial institutions, loans to non-financials corporate clients,
loans to retail and small business customers, and loans to sovereigns, central banks and PSEs
65%
Unencumbered residential mortgages with a residual maturity of one year or more and with a
risk weight of less than or equal to 35%
Other unencumbered loans not included in the above categories, excluding loans to financial
institutions, with a residual maturity of one year or more and with a risk weight of less than or
equal to 35% under the Standardised Approach

85%
Other unencumbered performing loans with risk weights greater than 35% under the
Standardised Approach and residual maturities of one year or more, excluding loans to
financial institutions
Unencumbered securities that are not in default and do not qualify as HQLA including
exchange-traded equities
Physical traded commodities, including gold

100%
All assets that are encumbered for a period of one year or more
Derivatives receivable net of derivatives payables if receivable are greater than payables
All other assets not included in the above categories, including non-performing loans, loans to
financial institutions with a residual maturity of one year or more, non-exchange-traded
equities, fixed assets, pension assets, intangibles, deferred tax assets, retained interest,
insurance assets, subsidiary interests, and defaulted securities
Source: BIS

Figure 45b: Composition of off-balance sheet categories and associated RFS factors.
RSF Factor Components of RSF category
5% of the
currently
undrawn portion
Irrevocable and conditionally revocable credit and liquidity facilities to any client
National
supervisors can
specify the RSF
factors based on
their national
circumstances
Other contingent funding obligations, including products and instruments such as :
Unconditionally revocable credit and liquidity facilities
Trade finance-related obligations (including guarantees and letters of credit)
Guarantees and letters of credit unrelated to trade finance obligations and
Non-contractual obligations such as
- potential requests for debt repurchases of the bank's own debt or that of related
conduits, securities investments vehicles and other such financing liabilities
- structured products where customers anticipate ready marketability, such as
adjustable rate notes and (VRDNs)
- managed funds that are marketed with the objective of maintaining a stable value
Source: BIS
Scope Ratings

February 2014 60 / 65

Interestingly the proposed RSF factors in the new consultation paper seem slightly more accommodating, in
particular the lower RSF factor for unencumbered loans to retail and small business customers (going from 85%
to 50% for retail and SME customers not qualifying for a 35% or lower risk-weight).
Overall, as a ratio to forecast, the NSFR is manageable to the extent that the different components can be
identified. The analyst will have to make assumptions on the degree of encumbrance of each portfolio (if this
information is not disclosed by the company) as well as on the stability of each deposit. With the enforcement
of the Bank Recovery and Resolution Directive (RRD), we believe that a more granular disclosure around
deposits will happen at one point or another.
Liquidity Coverage Ratio (LCR)
The LCR (Liquidity Coverage Ratio) is the only liquidity ratio of Basel 3 ready to be fully enforced. Its definition
has been fine-tuned in January 2013 and January 2014, and the ratio has been adopted by CRD IV (Articles
411 to 426 of CRD IV).
To quote the BCBS document Basel 3: The Liquidity Coverage Ratio and liquidity risk monitoring tools dated
January 2013: the objective of the LCR is to promote the short-term resilience of the liquidity risk profile of
banks. It does this by ensuring that banks have an adequate stock of unencumbered high-quality liquid assets
(HQLA) that can be converted easily and immediately in private markets into cash to meet their liquidity needs
for a 30-calendar day liquidity stress scenario.
Therefore this ratio can be defined as:
Stock of HQLA / Total net cash outflows over the next 30 calendar days 100%
In which Total net cash outflows over the next 30 calendar days = Total expected cash outflows Min
(total cash expected cash inflows; 75% of total expected cash outflows).
The HQLAs are liquid assets that are supposed to be liquid in markets during a time of stress and ideally be
central bank eligible.
Calculation of the numerator and the denominator
There are two categories of assets that can be included in the stock of High Quality Liquid Assets. Level 1
HQLAs can be included without limit. Level 2 can only comprise up to 40% of the stock.
Supervisors may also choose to include within Level 2 an additional class of assets (Level 2B assets). If
included, these assets should comprise no more than 15% of the total stock of HQLAs. They must also be
included with the overall 40% cap on level 2 assets.
The 40% cap on Level 2 assets and 15% cap on Level 2B assets should be determined after the application of
required haircuts, and after taking into account the unwind of short-term securities financing transactions and
collateral swap transactions maturing within 30 calendar days that involve the exchange of HQLAs.
Instead of copying the definition from the master Basel document, we have decided to reproduce the Annex 4 of
the paper entitled Basel 3: The Liquidity Coverage Ratio and liquidity risk monitoring tools in Figures 46a, b, c
and d.
Scope Ratings

February 2014 61 / 65

Figure 46a: definition of Stock of HQLA
Item Factor
A. Level 1 assets:
Coins and bank notes 100%
Qualif ying marketable securities f rom sovereigns, central banks,
PSEs, and multilateral development banks
Qualif ying central bank reserves
Domestic sovereign or central bank debt f or non-0% risk-weighted sovereigns
B. Level 2 assets (maximum of 40% of HQLA):
Level 2A assets
Sovereign, central bank, multilateral development banks, and PSE 85%
assets qualifying f or 20% risk weighting
Qualif ying corporate debt securities rated AA- or higher
Qualif ying covered bonds rated AA- or higher
Level 2B assets (maximum of 15% of HQLA)
Undrawn value of restricted-use committed liquidity f acility f rom Central Banks (RCLF) 100%
Qualif ying RMBS 75%
Qualif ying corporate debt securities rated between A+ and BBB- 50%
Qualif ying common equity shares 50%
Stock of HQLA

Source: BIS
Scope Ratings

February 2014 62 / 65

Figure 46b: Definition of cash outflows (part 1)
A. Retail deposits:
Demand deposits and term deposits (less than 30 days maturity) :
Stable deposits (deposit insurance scheme meets additional criteria) 3%
Stable deposits 5%
Less stable retail deposits 10%
Term deposits with residual maturity greater than 30 days 0%
B. Unsecured wholesale funding:
Demand and term deposits (less than 30 days maturity) provided by small business customers:
Stable deposits 5%
Less stable deposits 10%
Operational deposits generated by clearing, custody and cash management activities 25%
Portion covered by deposit insurance 5%
Cooperative banks in an institutional network (qualifying deposits with the centralised institution) 25%
Non-financial corporates, sovereigns, central banks, multilateral development banks, and PSEs 40%
If the entire amount fully covered by deposit insurance scheme 20%
Other legal entity customer 100%
C. Secured funding:
Secured funding transactions with a central bank counterparty or backed by Level 1 assets with any counterparty 0%
Secured funding transactions backed by Level 2A assets, with any counterparty 15%
Secured funding transactions backed by non-Level 1 or non-Level 2A assets, with domestic sovereigns, multilateral
development banks, or domestic PSEs as a counterparty
25%
Backed by RMBS eligible for inclusion in Level 2B 25%
Backed by other Level 2B assets 50%
All other secured funding transactions 100%
Cash Outflows
Source: BIS
Scope Ratings

February 2014 63 / 65

Figure 46c: Definition of cash outflows (part 2)
D. Additional requirements:
Liquidity needs (eg collateral calls) related to f inancing transactions, derivatives and other contracts 3 notch downgrade
Market valuation changes on derivatives transactions (largest absolute net 30-day collateral f lows
realised during the preceding 24 months)
Look back approach
Valuation changes on non-Level 1 posted collateral securing derivatives 20%
Excess collateral held by a bank related to derivative transactions that could contractually be called at
any time by its counterparty
100%
Liquidity needs related to collateral contractually due from the reporting bank on derivatives
transactions
100%
Increased liquidity needs related to derivative transactions that allow collateral substitution to non-
HQLA assets
100%
ABCP, SIVs, conduits, SPVs, etc:
Liabilities from maturing ABCP, SIVs, SPVs, etc (applied to maturing amounts and returnable assets) 100%
Asset Backed Securities (including covered bonds) applied to maturing amounts 100%
Currently undrawn committed credit and liquidity f acilities provided to:
retail and small business clients 5%
non-financial corporates, sovereigns and central banks, multilateral development banks, and PSEs
10% f or credit
30% f or liquidity
banks subject to prudential supervision 40%
other financial institutions (include securities firms, insurance companies)
40% f or credit
100% f or liquidity
other legal entity customers, credit and liquidity f acilities 100%
Other contingent funding liabilities (such as guarantees, letters of credit, revocable credit and liquidity
f acilities, etc)
Trade f inance
Customer short positions covered by other customers collateral
National discretion
0-5%
50%
Any additional contractual outf lows 100%
Net derivative cash outf lows 100%
Any other contractual cash outflows 100%
Total cash outflows

Source: BIS
Scope Ratings

February 2014 64 / 65

Figure 46d: Definition of cash inflows
Maturing secured lending transactions backed by the following collateral :
Level 1 assets 0% 0%
Level 2A assets 15%
Level 2B assets
Eligible RMBS 25%
Other assets 50%
Margin lending backed by all other collateral 50%
All other assets 100%
Credit or liquidity f acilities provided to the reporting bank 0%
Operational deposits held at other f inancial institutions (include deposits held at centralised institution of network
of co-operative banks)
0%
Other inflows by counterparty:
Amounts to be received f rom retail counterparties 50%
Amounts to be received f rom non-financial wholesale counterparties, f rom transactions other than those listed
in above inf low categories
50%
Amounts to be received f rom f inancial institutions and central banks, from transactions other than those listed in
above inf low categories.
100%
Net derivative cash inf lows 100%
Other contractual cash inf lows National discretion
Total cash inflows
Total net cash outflows = Total cash outflows minus min [total cash inf lows, 75% of gross outf lows]
LCR = Stock of HQLA / Total net cash outflows
Cash Inflows
Source: BIS
The definitions reproduced here are self-explanatory. Versus the situation two years ago, the definitions related
to cash inflows and cash outflows are now much clearer. In its most recent document, it is interesting to note
that BCBS makes certain Central Banks facilities eligible as Level 2B assets (while previously these facilities
could only be used by jurisdictions with insufficient HQLAs to meet the needs of the banking system).
Despite the increased precisions, the LCR remains as difficult to forecast as it ever was. However, we believe that the
analyst can arrive at a good proxy for HQLAs, considering that not many assets are eligible for HQLA status. Usually
most of these assets are willingly disclosed (except for RMBS and covered bonds which are Level 2 assets).
As a result of the above, Scope ratings has included in its Bank Rating Methodology published last week some
ratios based on Level 1 HQLAs. Because of the way Level 2 assets are capped, Scope can estimate a
maximum level of HQLAs that can be defined as:
Max. Total HQLAs = Level 1 HQLAs / 60%
Based on the fact that Total HQLAs = Level 1 Assets + Level 2 Assets.
Using Level 2 Assets under the constraint of the 40% cap, we can rewrite the equation above as:
Max. Total HQLAs = Level 1 Assets + (40% of Total HQLAs)
Or 60% Max. Total HQLAs = Level 1 Assets.
While disclosure improves on the LCR and before there is more clarity on the NSFR, Scope ratings will use
more traditional liquidity ratios such as comparing liquid assets to total wholesale funds or short-term
wholesale funds.
Scope Ratings

February 2014 65 / 65

Scope Ratings GmbH
Lennstrae 5
10785 Berlin
T: +49 (0)30 27891-0
F: +49 (0)30 27891-100
Service: +49 (0)30 7891-300

Hamilton House, suites 209 and 345
Mabledon Place, London WC1H 9BB
T: +44 207 554 8638 / +44 207 554 8615

[email protected]
www.scoperatings.com

Forecasting Bank Financials Methodology

Uploaded by

Copyright:

Available Formats

Forecasting Bank Financials Methodology

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Forecasting Bank Financials Methodology

Uploaded by

Copyright:

Available Formats

Contacts

Jacques-Henri Gaulard Sam Theodore Pauline Lambert

You might also like