I have only taken LI so far…does anyone know if the topics that grover33 talks about will be addressed in LII or LIII? If not, that’s very frustrating.
Not really - that’s the point. That said, I am pretty surprised you have just taken LI. I think you can learn this stuff pretty easily in some other ways.
I’m not worried about learning it, given it is what I do on a daily basis. I am just making the comment/observation that it is frustrating that people can get the CFA charter and not have a clue about this market. I’d say that this is a major flaw in the curriculum.
mib20 Wrote: ------------------------------------------------------- > I’m not worried about learning it, given it is > what I do on a daily basis. I am just making the > comment/observation that it is frustrating that > people can get the CFA charter and not have a clue > about this market. I’d say that this is a major > flaw in the curriculum. People have always and will always complain about this. The fact of the matter is the curriculum will always be a mile wide and an inch deep, and I really don’t think that’s a bad thing. There will never be a designation from which you emerge completely prepared to both run a leveraged long/short credit fund and manage an individuals fixed/equity portfolio. Does learning triangular arbitrage really give me enough knowledge about currency markets? Let’s step back for a second and think about the business of investment management. Where are most people actually working? Now do you want a designation that appeals to a broad swath of individuals, thus making it widely known and respected, or have it be highly applicable to a select group, but toil in obscurity like the CAIA? Now let’s also consider, and I realize this is unlikely, but maybe, just maybe, you think these topics are more important because you work with them on a daily basis and it is you, and not the rest of us, who could use some diversification of knowledge. FYI, you’ll get to behavioral finance on L3. Let me know if any of that applies to this conversation.
I’m not suggesting that a charterholder be able to recite Duffie and Hull, but I do think that having knowledge of CDS that is more than, “I have heard that this silly acronym has caused a lot of pain in today’s market and I heard it relates to the credit markets”, is appropriate for a designation that is so highly regaded in the profession of financial mgmt and analysis. I am pretty sure that I could find at least 5 questions on topics that were repeated elsewhere in the exam. I am not questioning whether or not the CFA does a good job at giving candidates and chartholders a wide range of skills to be succesful in this business, but rather, they should get a little bit more up to speed with credit type of vehicles. In terms of your comment regarding diversification of knowledge, it is almost laughable. When looking at distressed credit, you not only need to bea good equity analyst for hedging purposes and because you will effectively be an equity holder if the company defaults, but also you need to understand the fundamentals that drive credit valuation. So I would say that the tools used in credit are a bit more diversified that those who are purely in the equity markets.
First of all, don’t equate what you saw on this year’s L1 exam to “the curriculum”. The questions vary widely from year to year. Second, there are 2 more levels to come. Given the prereqs for the exam, in depth coverage of CDS on L1 seems pretty inappropriate. Third, since we’re all good analysts, let’s look at the primary sources. I’m looking at the L3 books right now, and there’s a whole study session named “Portfolio Management of Global Bonds and Fixed Income Derivatives”. In fact, fixed income investing is given 2 full study sessions versus just one for equities. Risk management also gets 2 study sessions. CDS are given 2 pages, with a decent rundown of what they are, including a schematic of the flows and a question set. I mean, would you like them to get into pricing? If you can do that on your TI, hats off to you. I’m just trying to say the curriculum’s got a lot to cover and does a pretty good job, and CDS are pretty far down the list of areas I would expand focus on.
this thread went from interesting to lame in a hurry (my opinion). grover, the JPM Credit Derivatives handbook, which is really a CDS primer (and a very good one, quite detailed at 160 pages or so) has a good, fairly simple section on tranched index products.
agreed…let’s get back to more talk about CDS pricing theory discussion
Thanks nodge. Ill go get that now.
Bloomberg story about LIFFE becoming a clearing house for CDS. That could go some way to alleviating concerns about counterparty risk. http://www.bloomberg.com/apps/news?pid=20601087&sid=atz_KAYTneMQ&refer=home
Hmmm, no replies. It’s almost as if there’s little interest in CDS pricing theory among CFA candidates. /sorry, couldn’t help myself.
I think a centralized exchange with publicly-disclosed margin requirements is an almost inevitable outcome for the CDS market given its growing size and importance. Should address a number of concerns over this product, mainly counterparty risk but also price transparency. I imagine the main hurdle is this would (I think) cut into dealer profitability. So mib, I understand the basics of CDS pricing. Can you explain, conceptually as opposed to mathematically, convexity in CDS spreads? Basically what is going on in the relationship of spread, recovery rate, and probability of default such that a rising spread leads to a lower DV01? Is it just because at higher spread levels the incremental default probability of a 1 bp move is much smaller on a proportional basis, and therefore the annuity is less sensitive to the 1 bp shock as hazard rate are increasing at a declining rate? It’s just one of those concepts that doesn’t “click” for me even though I can clearly observe it while playing with a simple CDS pricing model.
Ok, well I will start with a simple math equation and then show you why this conceptually makes sense. This will all assume that we are dealing with a 1yr CDS contract. So, generally speaking, the fair amount that you will pay for default protection should be equal to the sellers loss time the probability that he/she actually suffers that loss. Thus we get the equation: CDS Spread = (1-recovery)*probability of default Using this equation we can use readily aailable spreads and an assumption for recovery in order to back into probability of default. To do so, we rearrange the terms to read: probability of default = CDS Spread/(1-recovery) Thus, at tighter spread levels, changes in recovery will have very little impact on probability of default. However, as the spread becomes increasingly wider, recovery assumptions will become much more of a relevant factor. Now that you see how this equation works, you can see why higher spreads will lead to higher probabilities of default as well as the impacts that recovery assumptions will play in probability of default calculations. This is why increasingly higher spreads will continue to decrease your DV01 (keeping recovery constant of course). I hope this helps.
Uh, yeah I’m familiar with those relationships, I guess I’m just looking to close the loop between this sentence: “Now that you see how this equation works, you can see why higher spreads will lead to higher probabilities of default as well as the impacts that recovery assumptions will play in probability of default calculations.” and this sentence: "This is why increasingly higher spreads will continue to decrease your DV01 (keeping recovery constant of course). " The first one is obvious, but the link between the first and second is, at least to me, not intuitive.
Just use the equation that I gave you: CDS Spread = (1-rr)*pd Try to understand this equation and you will be able to understand why these relationships make sense. If you are unclear why the above equation works, just think about the fair value that you would be willing to accept for taking on the risk of insuring default risk. You would want to be compensated for the potential loss you could experience weighted by the probability that you actually experience that loss.
To keep the tread going, I would add some inputs and may duplicate some of the things discussed already: The key inputs to the price of CDS: 1. probability of degault of the reference entity and protection seller 2. Correlation between the reference entity and protetction seller 3. Joint probability of default of the reference entity and protetction seller 4. maturity of swap 5. Expected recovery value The trade can be replicated via 1. Purcahse a cash bond 2. Pay fixed on a swap (same maturity as the cash bond) receive LIBOR 3. Finance the bond purchase in the repo market, pledge the bond as collateral
Ok mib, I did some reading and I see what I was missing. Maybe it was implied by what you’re saying, but the (obvious) feedback mechanism is that increasing spreads (and therefore increasing default probabilities) lower the value of your risky annuity, hence making each incremental increase in spread less detrimental to your M2M (assuming you’re a protection seller). And that impact is, also somewhat obviously, more dramatic for longer maturities as the hazard rates are compounded. Thanks, you posted a lot of good stuff on this thread.
No problem at all. Glad to see that you were able to get a decent handle on this stuff.
Complete agree, mib20, thanks for the input.