08' practice exam relevance?

[hsuwang]

Hello David,

As I'm working through GARP's 2008 Practice Exam, it is reasonable to bump into non-relevant questions to 09's material, but I'm not so sure if I should be worried about them? For example, in the 08' practice, it tested RAROC concept with the given information: after-tax profit, Net book market value, Weekly volatility, and Tax rate (which we did not cover this year), but since RAROC sounds so familiar in regard to this year's curriculum, I'm not sure if I should dig deeper into this material...

Thanks.


reference: GAPR 2008 practice exam II #32

 

[ajsa]

Hi David,

I also wonder if we need consider tax for RORAC? it looks like the current fomula is pre-tax?

thanks.

 

[David Harper CFA FRM]

Hi Jack & asja,

On the specific question, I think it's flawed (it was on my error list to GARP last year): we've never reviewed (in any source) an after-tax RAROC; and the reference (de Servigny) uses a pre-tax version consistent with the assiged Crouhy (so even those using de servigny were confused by this). Beyond that, I am unsure whether it's even correct: why is the denominator (VaR) tax-adjusted? True, it's ratio consistent (after-tax in numerator and denominator) but VaR/EC are not capital resources, I am unclear on why the VaR would be tax-adjusted (e.g., EVA will inform the denominator with WACC, and the debt gets a tax shield, but here the entire VaR is "tax shielded"?)
... I have customers who know more than me about actual RAROC practice ... I would love to know if the denominator in an after-tax RAROC should be multipled by (1-tax) and what is the rationale (it's not like it's debt with a tax shield?)? I would think after-tax RAROC = after-tax return / (VaR | EC but not after-tax)?

@Jack: in regard to whether you should dig deeper on RAROC: it's not necessary, you should only need the one definition from Crouhy
@Asja: no, you should not need to consider tax ... the 2009 exam should be flushed of this distraction

more generally, it's amazing how we can learn even from bad questions, sometime i think i learn more from bad questions! For example, this quesiton does remind us that we don't need to get stuck on EC as the risk metric in the denominator. It is very characteristic of the FRM to "switch" the EC with VaR, because it's okay to do that, but it will trip up a pure memorizer. While a better question would clarify whether the ratio is net of EL (i.e., the annual time horizon implies a non-zero EL, so the VaR can either exclude or include EL. Our Crouhy method excludes EL), it is still perfectly *fine* to use VaR in the denominator.
...so it's maybe not too hepful, but I think using historical questions can be very helpful if we are a bit discriminating and skeptical
...it won't help too much for 2009, but i'll add a section to the forum which parses good vs. bad old exam questions, partly b/c I want to be able to point GARP to the flawed questions

thanks, David