RDP 9409: Default Risk and Derivatives: An Empirical Analysis of Bilateral Netting 7. Coverage

Marianne Gizycki and Brian Gray

December 1994

This section presents some evidence on the extent to which credit exposure falls when moving from a non-netting to a netting environment.

The reduction in replacement cost (current exposure) depends upon the composition of each bank's portfolio; in particular the value of out-of-the-money contracts. For banks 1 to 7, we estimate that moving from calculating current exposure on a gross basis to a net basis, will reduce replacement cost by an average of 33 per cent (see Table 15). This assumes that netting agreements are concluded with all counterparties who conduct two-way business.

Table 15: Percentage Change from Non-Netted to Netted Exposure
    The gap between total
exposure and current
exposure:
  Maximum exposure Average exposure Capital for current exposure Maximum exposure Average exposure
Bank 1 −16.89 −39.42 −20.88 −6.53 2.15
Bank 2 −18.66 −21.44 −25.13 −7.08 −31.89
Bank 3 −36.11 −32.46 −14.61 −41.46 −49.10
Bank 4 −35.12 −39.72 −70.04 31.17 −123.89
Bank 5 −13.77 −19.49 −43.92 27.75 −200.28
Bank 6 −8.34 −9.13 −7.78 −10.21 −6.45
Bank 7 −21.55 −23.69 −37.59 8.77 −70.99
Total −20.98 −27.53 −33.13 1.02 −44.59

It is quite possible that, in a netting environment, the reduction in current exposure may be larger (both in proportional and absolute terms) than the reduction in maximum future exposure, which implies that potential exposure may in fact increase. Consider the case of a portfolio of contracts with negative market value overall, but which contains some positively valued contracts. The gross current credit exposure will have some positive value, while the net current exposure will be zero. Hence, adoption of netting will result in a 100 per cent reduction in current credit exposure. However, it is quite possible that in both the gross and net environments, future potential exposure will take on some positive value. While the gross potential exposure may be considerably higher than the net potential exposure, so long as net potential exposure is positive, the proportionate reduction in potential exposure when moving from gross to net will be less than 100 per cent.

Table 15 shows that the average increase in the gap between maximum exposure and the capital held for current exposure is small, at just one per cent. However, for individual banks the gap can increase by as much as 30 per cent. Reflecting this, Table 16 shows that the coverage of potential exposure by the Basle add-on is little different, overall, between the netted and non-netted scenarios. Table 16 also demonstrates that while the Basle measure of credit exposure covers only a proportion of maximum credit exposure (a reflection, in part of the comparatively severe assumptions behind this measure of exposure) it more than adequately covers average credit exposure.

Table 16: Coverage
  Percentage of potential exposure covered by the Basle add-ons Percentage of total exposure covered by the Basle credit equivalent
Non-netted Netted Non-netted Netted
Max exposure Avg exposure Max exposure Avg exposure Max exposure Avg exposure Max exposure Avg exposure
Bank 1 10.99 100.00 11.76 100.00 76.27 190.63 72.45 248.41
Bank 2 10.98 100.00 11.82 100.00 65.39 157.44 63.92 159.33
Bank 3 7.78 29.18 13.28 57.34 33.36 80.89 36.35 83.36
Bank 4 18.17 100.00 13.85 111.18 68.73 163.99 39.91 102.49
Bank 5 10.83 100.00 8.48 58.04 59.54 118.84 42.96 91.83
Bank 6 20.78 100.00 23.14 100.00 145.02 380.73 82.69 219.00
Bank 7 12.60 100.00 11.59 100.00 65.40 141.59 57.59 128.18
Total 11.98 100.00 11.86 100.00 64.43 148.81 59.93 150.90