Docstoc

The vdp-Curve From Pfandbrief Yield to Mortgage Interest

Document Sample
The vdp-Curve From Pfandbrief Yield to Mortgage Interest Powered By Docstoc
					     The vdp-Curve:
     From Pfandbrief Yield to
     Mortgage Interest




     Over the course of the financial market crisis, the spread differentials between Mortgage
54   and Public Pfandbriefe had risen considerably. This was the reason for the vdp to split up
     the vdp Pfandbrief Curve, which had included both types: The vdp-Curve (Mortgage Pfand-
     brief) and the vdp-Curve (Public Pfandbrief) were developed. The Mortgage Pfandbrief
     type is designed to provide Pfandbrief Banks with a reliable basis for their calculation of
     mortgage loan terms and to help clients to understand the conditions of these loans. The
     new curve has been launched (and published since) 1 February 2009 – a good reason to
     explain the concept. This article explains the connections between Pfandbrief yield and
     mortgage loan interest rate in particular.



     Christian Fischer | Deutsche Hypothekenbank (Actien-Gesellschaft)
     Bodo Winkler | Association of German Pfandbrief Banks (vdp)
     Make Two from One


For quite some time after the current financial crisis initially hit the US subprime market
in July 2007 it looked like the Pfandbrief might completely escape the waves of financial
upheaval spreading across global markets. Not until September 2008 did the primary market
for Pfandbriefe succumb, only for a short time but with full force. It then became apparent that
the Pfandbrief, the last bastion, had been hit directly by the financial crisis. Up until 2007 the
spreads on Pfandbriefe had been steadily narrowing, yield differentials on issues from differ-
ent issuers had almost disappeared, and Mortgage and Public Pfandbriefe had nearly identical
levels of return. But some parameters of the Pfandbrief market began changing in the second
half of 2007. Jumbo Pfandbriefe could no longer be placed at any given time, but only during
certain windows of time. The bid-offer-spreads in market making for Jumbo Pfandbriefe
doubled, and eventually even tripled. And the market started differentiating more again
between the various issuers and also between Mortgage and Public Pfandbriefe. In spring
of 2008, the differences between yields on instruments secured by mortgage loans and those
secured by public sector loans from the same issuers and of the same duration exceeded
10 bp for the first time in a long time. Previously this yield differential had typically remained
around 1 to 2 bp. This development got the vdp working group “Quality Assurance of the
vdp-Pfandbrief Curve” thinking about differentiating the two types of Pfandbriefe represented
in the vdp-Pfandbrief Curve. The Pfandbrief and Capital Markets Committee decided in favour
of a differentiation at its meeting of 27 July 2008 and preparatory work got under way. On
1 February 2009 the new vdp-Curve (Mortgage Pfandbrief) went “live”. It solely reflects the          55

funding levels of Pfandbrief Banks with respect to Mortgage Pfandbriefe, while Public Pfand-
briefe are reflected in a separate curve, the vdp-Curve (Public Pfandbrief). Both curves are
combined to form the vdp-Pfandbrief Curve (old), which is calculated as the averages of the
two new curves weighted by the outstanding Pfandbrief volumes.



     Function of the vdp-Curve (Mortgage Pfandbrief)


As the differences between Mortgage and Public Pfandbriefe became more pronounced, it
became necessary to work with two different curves, in particular in view of the function the
“old” vdp-Pfandbrief Curve had for the mortgage lending business conducted by the vdp
member banks. The Pfandbrief Curve was launched in April 2003 by vdp's predecessor, the
VDH, in order to give member banks a basis for calculating bank-internal transfer prices. It
was also intended as a sales support instrument and an argumentation aid for use with clients.
How the vdp-Curve (Mortgage Pfandbrief) now fulfils theses functions and how it is used by
Pfandbrief Banks in the value creation chain between the issue of a Mortgage Pfandbrief and
design of a mortgage loan is discussed here in the sections “The basis of cost-of-funds calcu-
lations by the Pfandbrief Banks” and “From Pfandbrief yield to lending interest rate – the mar-
gin calculator of Deutsche Hypo”. Other functions, such as acting as a benchmark for bond
pricing or as an indicator of appropriate price levels in the Pfandbrief market for investors,
should not be excluded, but are not the focus of these considerations.
     The vdp-Curve (Mortgage Pfandbrief): From Pfandbrief Yield to Mortgage Interest




     To serve these purposes, at the time it was regarded as essential to create a yield curve that
     would reflect the totality of the Pfandbrief market as far as possible. The mortgage banks com-
     prising the VDH therefore decided that all the member banks, irrespective of whether their
     business was concentrated on public sector or mortgage financing or equally spread across
     both segments, would contribute their individual funding mix to the calculation of the curve.
     Thus, both Pfandbrief types were included, as well as traditional bearer and registered Pfand-
     briefe and Jumbo issues. Because yield differentials among the various Pfandbrief segments
     were minimal, this mixing did not negatively influence the curve’s effectiveness as a tool for
     lending business.



             Structure of the vdp-Curve (Mortgage Pfandbrief)


     The vdp-Pfandbrief Curve was conceived in 2003 as an interest rate curve covering maturities
     from one to ten years. Maturities longer than ten years also play an important role in mortgage
     lending business, but the Mortgage Bank Act stipulated that the right of early repayment could
     not be excluded for Mortgage Pfandbriefe, in contrast to Public Pfandbriefe. Thus, for longer
     maturities, options would have had to be taken into account for Mortgage Pfandbriefe, which
     would have resulted in pronounced discrepancies in the yields for Mortgage Pfandbriefe as
     compared to Public Pfandbriefe. Since a single curve was to serve for both types of Pfandbrief,
     it was therefore decided to exclude the long maturities, in the interest of preserving trans-
56   parency. When the Pfandbrief Act entered into force on 19 July 2005 and the requirement
     stipulated in § 8 par. 2 Mortgage Bank Act was repealed, the maturity range covered by the
     curve was extended to 15 years, the same standard is now applied to the vdp-Curve (Mortgage
     Pfandbrief).

     Yield
     5.000
                                                                                                                                4.844
     4.750

     4.500                                                                                                          4.623
                                                                                                      4.445
     4.250                                                                                   4.342
                                                                                     4.217
     4.000                                                                   4.070
                                                                     3.883
     3.750
                                                             3.661
     3.500
                                                     3.344
     3.250

     3.000
                                             2.944
     2.750

     2.500
                                 2.416
     2.250

     2.000

     1.750
                     1.773                                                                                       Fixing as of June 12, 2009
     1.500
                 1           2           3       4       5       6       7       8       9       10      11    12      13     14        15

                                                                                                       Time remaining to maturity (years)
    A special feature of the vdp-Curve (Mortgage Pfandbrief) and its sister index, the vdp-
Curve (Public Pfandbrief), along with their forerunner, the vdp-Pfandbrief Curve, is that the
information from all the participating banks is weighted equally. This approach was decided
when the curve was conceived so as to motivate smaller issuers to also participate. Moreover,
in a test phase preceding the initial launch of the VDH-Pfandbrief Curve in 2003 showed that
weighting the information from the individual banks according to their share of outstanding
Pfandbriefe did not alter the results. This paved the way to all the VDH members agreeing to
participate in the single-curve model and ultimately identifying with it. In contrast, the newer
vdp-Curve (Mortgage Pfandbrief) is calculated only from information provided by institutions
for whom real estate financing and the associated refinancing via mortgage Pfandbriefe is a
regular core business. At present there are a total of 20 Pfandbrief Banks, ranging from clas-
sical mortgage banks such as Deutsche Hypothekenbank (Actien-Gesellschaft) to universal
banks like Deutsche Postbank to Landesbanken like BayernLB.



         Calculation and Publishing of the vdp-Curve (Mortgage Pfandbrief)


On every banking day, every participating institution reports for each maturity the average
funding costs for its individual mortgage funding mix expressed as a spread versus mid-swaps.
The spread is based on currently attained or respectively currently attainable funding levels on
the primary market. Structured features are not included here, and if they are normally con-
tained in an institution’s funding mix, then they have to be factored out.                                                          57

    The spread reports are forwarded via Reuters’ contributor pages to Moosmüller & Knauf.
Moosmüller & Knauf calculate the average spread per maturity on the basis of the reports
received by 10.30 a.m. CET. Immediately following ISDA swap fixing, the vdp-Curve (Mortgage
Pfandbrief) is fixed on the basis of these average spreads, and this fixing curve is then pub-
lished on Reuters and on the vdp website.



vdp-PFANDBRIEF CURVES DURING THE COURSE OF THE TRADING DAY


                                       Calculation of average   Calculation of vdp-Pfandbrief         Use of just-calculated
             Banks report                                                Curve (old)             average spreads for vdp-Curve
                                       spreads for vdp-Curve
                current                                                                              (Mortgage Pfandbrief)
                                       (Mortgage Pfandbrief)
          spreads to Reuters                                                                    and vdp-Curve (Public Pfandbrief)
                                           and vdp-Curve
             by 10.30 a.m.                                                                           in the real-time curves
                                         (Public Pfandbrief)        Synchronized fixing
                                                                    of Pfandbrief Curves




  9.00                         10.30      11.00                                                                            17.00




                                                                   Update of vdp website,
                                       ISDA swap fixing
                                                                 release of curve data to the
                                          11.00 a.m.
          Start of                                                          media                              Stop of
     real-time curves                                                                                     real-time curves
     The vdp-Curve (Mortgage Pfandbrief): From Pfandbrief Yield to Mortgage Interest




     Reuters also publishes the “vdp-Real Time Yield Curve (Mortgage Pfandbrief)”, which results
     from the addition of the average spreads that are calculated once daily plus the real-time swap
     curve. The fixed values serve in particular as the basis on which lending conditions are negoti-
     ated with borrowers. They are published via Reuters and the vdp website, as well as by Bloom-
     berg, Handelsblatt and the monitor applications of vwd. With the real-time curve, the two new
     vdp Curves fulfil the requirements of modern, always current indexes. They are updated con-
     tinuously between 9 a.m. and 5 p.m.



          The Basis of Cost-of-Funds Calculations by the Pfandbrief Banks


     The vdp-Curve (Mortgage Pfandbrief) is supposed to serve primarily as a basis for calculating
     the cost of funds in commercial real estate lending business. Given the various possibilities of
     structuring real estate loans in day-to-day business, e.g. with respect to individual contractual
     elements such as repayment models, interest payment frequency, forward conditions, potential
     cash flow structuring (funds paid out in accordance with building progress) and amount of
     the collateral portion, if an adequate response is to be ensured, it is not enough to update the
     vdp Curve only once a day in banks’ IT systems. Rather, acquisition and credit officers need to
     have constant access to market-consistent cost rates for funds. At most Pfandbrief Banks, this
     is the job of the treasury or asset-liability management.
         The interest rate communicated by the internal systems of the Pfandbrief Banks is ideally
58   a fixed interest rate that assumes the mortgage loan in question is a fully coverable loan with
     interest due annually. For a loan with a loan-to-value ratio of 60% of the mortgage lending
     value or less, the vdp-Curve can be applied to the entire amount of the loan because it is fully
     subject to the Pandbrief cover. Often the amount of the loan exceeds the 60% loan-to-value
     limit, so only part of the loan is coverable. The portion that is in excess cannot be refinanced
     through Pfandbriefe. Other funding instruments have to be used instead. In the case of classi-
     cal mortgage banks without deposit business, usually uncovered bank bonds are used for this
     purpose.
         The Pfandbrief Banks calculate loans like this that are not purely first rank using a composite
     rate, or they separate the loan into two portions, one coverable and one not coverable. In the
     latter case, the basis of the cost-of-funds calculation is different for the two portions of the loan:
     the vdp-Curve is applied to the coverable portion, and the refinancing costs for uncovered bank
     bonds to the uncovered portion. In addition, the contractual elements listed above may also be
     used in structuring the loan.
         Clearly, in carrying out these procedures, the banks have to process the “raw data” from
     the vdp-Curve further. For this purpose, Deutsche Hypo has devised a margin calculator, which
     is described below.
      From Pfandbrief Yield to Lending Interest Rate – the Margin Calculator
      of Deutsche Hypo


The interest rate shown on the vdp-Curve for a given maturity is based on the ISDA swap fix-
ing and the average of all the spreads reported by the Pfandbrief Banks for the maturity in
question, as explained above. The average premiums on the swap curve are stored in the mar-
gin calculator of Deutsche Hypo and can thus be used any time in loan calculations indepen-
dently of the absolute level of the ideal cost-of-funds rate. As a rule, it should always be clearly
defined with the borrower which cost-of-funds curve (vdp or swap) is to serve as reference.
Depending on which curve is used, different premiums may apply, but the total effective client
interest rate is the same in both cases.
      The margin calculator incorporates the cost-of-funds rates, including related spreads, in
the individual terms and conditions of the loan, using the 6-month EURIBOR as the basis, in
accordance with principles of the vdp-Curve and standard market practice. Loans with fixed
interest rates or rollover loans with half-yearly interest rate adjustments can be calculated in
this way without any further modifications. In connection with the variable underlying swap
curves, floating loans with quarterly or monthly terms of interest necessitate a conversion of
the requisite spreads through the incorporation of the underlying swap costs. Based on the
terms of interest specified, the margin calculator automatically calculates the correct premi-
ums. The swap curves for the 3-month and 1-month EURIBOR are available, too. The requisite
bank margins are calculated in accordance with basic loan-specific data (type of property, loca-
tion, type of loan, equity base). These details are stored in the margin calculator; the following                           59

tables do not take them into account, however.



Example: Financing; EUR 10 million, duration: 5 years, fixed interest rate,
loan-to-value ratio: 100%


  Loan in EUR million                           Spreads via ISDA swap fixing              Cost of funds           Client’s
                                                                                                                 effective
                                                 unsecured                     weighted
                                                                                                             interest rate
                                 vdp vs. swap      vs. swap       Total         average   ISDA swap fixing


     up to 60% loan-to-value 6       0.57%               –     0.57%
10                                                                              0.94%             2.95%            3.89%
     over 60% loan-to-value 4              –       1.50%        1.50%



  Loan in EUR million                            Spreads via vdp Curve                    Cost of funds           Client’s
                                                                                                                 effective
                                                                               weighted
                                                                                                             interest rate
                                     unsecured vs. vdp            Total         average   vdp Curve fixing


     up to 60% loan-to-value 6                      –          0.00%
10                                                                              0.37%             3.52%            3.89%
     over 60% loan-to-value 4                   0.93%          0.93%
     The vdp-Curve (Mortgage Pfandbrief): From Pfandbrief Yield to Mortgage Interest




     Calculating this way provides borrowers with a recognized means of comparing the refinanc-
     ing costs of lending institutions. By applying the vdp-Curve to the cover portion of the loan
     for different Pfandbrief Banks, the same or at least a similar cost-of-funds rate results for the
     client. Variations are thus the sole result of differences in the refinancing costs of the non-
     cover portion (e.g. via uncovered bank bonds). The difference between the cost-of-funds rate
     and the lending rate offered by the bank represents the bank’s lending margin.
           In discussions about conditions, the client therefore usually only negotiates on the
     amount of the lending margin charged. The use of the vdp-Curve (Mortgage Pfandbrief) by
     the Pfandbrief Bank means the client is in a position to evaluate competing offers. The cost-
     of-funds rates are also suitable for use in transactions with international clients who are not
     familiar with the vdp-Curve, because they also allow referencing to the swap curve.


          Benefit Derived from Conscientious Quality Assurance


     In order for the vdp-Curve (Mortgage Pfandbrief) to function within the Pfandbrief Banks as
     described here, it must be ensured that the curve reflects a yield level that is consistent with
     the market under all circumstances. The reason the forerunner curve, the vdp-Pfandbrief
     Curve, came into being had largely to do with a lack of confidence in the ability of the various
     established indexes at the time to accurately reflect reality. From its inception, the curve was
     therefore distinguished by the special quality assurance measures it applied. Moosmüller &
60   Knauf monitor on a daily basis the volatility and up-to-dateness of the spreads reported by the
     banks on the basis of specific mathematical criteria, and they immediately eliminate any ex-
     tremes they identify in order to prevent distortions of the curve. The main three measures they
     use for this purpose are:

     •	 	In	every	maturity	segment	of	the	vdp-Curve	(Mortgage	Pfandbrief),	the	lowest	and	the	
         highest spreads reported are omitted so as to prevent the influence of extreme values
         (outliers) on the curve.
     •	 	Data	that	differ	by	more	than	10	bp	from	the	data	reported	on	the	previous	day	are	not	
         used in the calculation of the average spread; instead the previous day’s value is taken for
         the calculation.
     •	 	In	the	event	that	a	given	institution	does	not	report	its	current	figures	to	Reuters	on	a	given	
         day by 10.30 a.m., Moosmüller & Knauf will use the data it has most recently received from
         that bank. This procedure will be applied for a maximum period of five banking days. As of
         the sixth day, spreads for the bank in question will no longer be included in the calculation.

     The results of this ongoing monitoring are summarised in an end-of-month quality report by
     Moosmüller & Knauf which also contains all the spreads reported by each participating bank
     for each maturity. On the basis of this quality report, the working group for “Quality Assurance
     of vdp-Pfandbrief Curves”, under the auspices of the vdp Pfandbrief and Capital Markets Com-
     mittee, reviews the development of the curve for the month in question. The working group
     responsible for quality assurance consists of three representatives from member institutions,
     one each from three different banks, and every six months one of these members is replaced
     by a new member from a different bank. This system of rotation allows every bank a turn in
     the working group. Usually the representative a bank nominates to participate in the working
     group is an employee who is active in the bank’s funding activities and therefore has an over-
     view of prices on the primary market.
In the monthly meetings of the quality assurance working group, the following topics are regu-
larly addressed:

•	 	Reporting	discipline	among	the	participating	lending	institutions:	If	a	bank	frequently	fails	
    to make a daily report, the quality assurance working group considers whether sanctions
    are necessary. In extreme cases, an institution may even be excluded from participating in
    the vdp-Curve (Mortgage Pfandbrief), which, however, has so far never been necessary. An
    improvement is usually achieved by making the institution aware of the irregularity of its
    reporting and requesting better discipline in future.
•	 	Plausibility	of	the	funding	levels	reported	by	the	banks:	If	the	working	group	has	any	
    doubts about the market conformity of the spreads reported by a given bank, it requests
    an explanation from the bank and may also request that the bank consider adjusting the
    spreads.
•	 	Pertinence	of	the	concept	behind	the	curve:	The	working	group	also	serves	as	a	forum	
    where the suitability and accuracy of the curve are discussed on an ongoing basis and any
    necessary changes can be initiated.


     Objectives Achieved – Thanks to Continuous Development Efforts


The role of the quality assurance working group to act as a “watchdog” ultimately resulted
in the old vdp-Pfandbrief Curve being split into the two new curves, vdp-Curve (Mortgage
Pfandbrief) and vdp-Curve (Public Pfandbrief). This developmental step means that the vdp-              61

Curve (Mortgage Pfandbrief) is suited to continue fulfilling its intended purpose for the mem-
ber banks even in the changed conditions of the Pfandbrief environment evolving out of the
financial crisis. To ensure that this remains so in future, the vdp working group will continue
to closely observe the Curve and initiate changes as necessary, in order to uphold the original
concept of the vdp-Curve (Mortgage Pfandbrief) as an instrument creating more transparency
in the lending business of the Pfandbrief Banks.
    The most important condition for the acceptance of the vdp-Curve (Mortgage Pfandbrief) is
that the funding levels it reflects are realistic values, i.e. it must be avoided that the Pfandbrief
Banks act as if they were in a race for the lowest refinancing rates. A natural guard against this
is the fact that banks referring to the Pfandbrief Curve in their lending business would effec-
tively put themselves at a disadvantage by behaving this way, because it would lead to lower
lending interest rates. On the other hand, the participating institutions would also suffer if refi-
nancing levels were too high, as they would be at a disadvantage against competitor institu-
tions that do not use the Curve.
    Naturally, some banks will tend to be above the average rates expressed in the Curve, and
others will tend to be below the average. But relative homogeneity of funding levels among the
participating institutions is important with a view to the original, central function of the Pfand-
brief Curve. Borrowers are given a reliable, constantly up-to-date and available basis for evalu-
ating lending conditions and can identify what portion of the interest rates a bank charges are
required to pay for the funding and what portion goes to cover the bank’s own costs and mar-
gin. The Curve is also a tool for the bank in explaining to clients any potential changes in the
interest rate between the time of the initial advisory discussion and the value date of the loan.
Furthermore, Pfandbrief investors have an actual market average at their disposal depicting
the cost-of-funding rates on the primary Pfandbrief market. Given that the vdp members repre-
sent approximately 98% coverage of the market with respect to new issues of Pfandbriefe, this
makes the Curve a legitimate benchmark.

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:6
posted:5/18/2012
language:English
pages:8
fanzhongqing fanzhongqing http://
About