[ Pobierz całość w formacie PDF ]
1998 20.5 18.2 15.6 20.5
1999 9.4 19.4 17.2 19.4
therefore investigated, using this index, in what follows. The course of the volatility is detailed
in Table 11.5. This was calculated for each calendar year with the help of the figures for the
last four, five and six year s indices, such that three, four and five quotients respectively were
available for the calculation (see equation (7.45)).
The highest volatility figure concerned is given in each line in the last column of Table 11.5.
These figures are used later in the course of this analysis to define a worst-case scenario. This
means that both the influences both of short- and long-term volatility are included in the later
calculations. It is of course debatable whether this is the right way to proceed. For the purposes
of this example, however, the volatility in each case per annum is determined in this way. The
figures in the table are shown graphically in Figure 11.2. It is striking that the volatility figures
show a clear peak at the beginning and at the end of the 1990s. The cause is in the first place
the sharp increase in the index figures and the subsequent sudden transition to a decline in
prices, and the uneven development in the second half of the 1990s.
Two assumptions vis-à-vis mortgages are made for the credit shortfall risk calculation. The
first assumption is that a multiple dwelling unit in 1985, i.e. well before the mortgage loan
crises at the beginning of the 1990s, is mortgaged at 80% of its saleable value. The amount of
the loan is left unchanged in subsequent years, without repayments, which leads to different
mortgage figures in each year corresponding to the course of the index. The credit shortfall
risk figures calculated in this way demonstrate how they developed for the bank over the years,
in line with the assumption that neither repayments of them, nor increases in them, took place.
The second assumption is that in each calendar year in each case a multiple dwelling unit
is mortgaged again to the extent of 80% of its saleable value. The credit shortfall risk figures
calculated in this way demonstrate what sort of risk the bank was running in each calendar
year concerned by financing to the extent of 80%.
Applications 137
Table 11.6 Credit shortfall risk development in % under the
assumptions
Year 1985 1986 1987 1988 1989 1990 1991 1992
Ann. 1 0.09 0.00 0.00 0.00 0.00 0.01 0.10 0.02
Ann. 2 0.09 0.08 0.01 0.00 0.03 3.27 3.91 1.82
Year 1993 1994 1995 1996 1997 1998 1999 2000
Ann. 1 0.00 0.00 0.00 0.99 0.88 1.07 0.28
Ann. 2 0.98 0.00 0.00 2.18 2.85 1.90 1.57
30%
25%
20%
4
5
15%
6
Max
10%
5%
0%
1985 1990 1995 2000
Year
Figure 11.2
As may be inferred from Table 11.6, the credit shortfall risks under the first assumption were
very small even during the crisis at the beginning of the 1990s. Under the second assumption,
however, they became plainly higher than previously at the beginning of the 1990s. As early
as 1990 the figure was so high that 80% mortgages in that year were no longer appropriate.
This situation repeated itself towards the end of the 1990s. The facts are portrayed graphically
in Figure 11.3. Annual volatility figures were used for the calculation. Appropriately higher
figures should be used when granting fixed mortgages over several years, which leads to credit
shortfall risks that are correspondingly even higher.
It becomes clear from Table 11.6 and Figure 11.3 that the mortgage loan crisis, based on the
annual index figures, was foreseeable from the model as early as the beginning of 1991, after
the index figure for 1990 came into existence. By using quarterly figures it may be assumed
that the reaction time could have been reduced even further.
It would be interesting, in the immediate future, to track down whether the credit shortfall
risks rates that have just risen again are leading to a second mortgage loan crisis in the near
Volatility
138 Risk-adjusted Lending Conditions
4%
1
3%
2%
1%
0%
2
-1%
1985 1990 1995 2000
Year
Figure 11.3
future, or whether the necessary safety precautions, on the basis of the experiences banks have
had, have been taken.
On the strength of this result the usefulness of the model proves itself, at least in qualitative
[ Pobierz całość w formacie PDF ]