Exam Feedback May 2021 Part 2 Exam Feedback

[Nicole Seaman]

We hope that everyone did well on the FRM Part 2 exam on Saturday! We would love to hear any feedback that you have about the exam. How did it go? Did you encounter unexpected questions? How was the experience with all of the COVID guidelines? Thank you in advance for any feedback you can provide!

IMPORTANT NOTE: THIS IS ONLY FOR THE PART 2 EXAM THAT WAS HELD ON SATURDAY, MAY 15TH. THE PART 1 EXAMS ARE STILL ONGOING SO WE CANNOT ALLOW PART 1 FEEDBACK TO BE POSTED IN THE FORUM UNTIL THE EXAMS ARE OVER ON THE 21ST.

 

[tami_92]

The exam was (at least for me) more difficult than expected for two reasons: 1. 80%-90% of it was qualitative with too much generic proposed answers, 2. extreme focus on certain topics (FRTB, EVT etc had repeated questions) with almost no testing on others ("poor diversification"), meaning that if your preparation wasn't that much specific on a given topic, you were likely to miss several points . While part I was more quant and therefore it was easier to understand if you did good or not, this part has left me with mixed feelings . Really don't know if it will be a pass or not: too many times I was left with two options on hand that could fit the right answer. Very challenging folks.

 

[FrmPart2AnonFeedback]

Overall I thought the exam was poorly written and I aced Part 1 so I have some credibility and am not just being sore about things.

Completely agree with tami_92 - topic selection focused disproportionately on some relatively obscure topics and not on core topics. I was glad I covered them, but I didn’t think it made for a fair test. Maybe on a 200 question test but not an 80 question one. And the questions/answers were very awkwardly written such that it was hard to tell if the choice contained a “gotcha” or was just sloppily written. As a native English speaker, I can’t imagine how much more difficult this would be for someone who isn’t. It further increased my respect for BT questions as they are difficult but very precisely written.

Given how qualitative the exam was I found it inappropriate to distill 1200 pages of dense reading into 80 vaguely worded MC about peripheral topics over four hours. It should have had more questions than Part 1.

BT exams were a good approximation for the difficulty. I got used to the well-worded questions in BT while the exam questions were poorly worded but in my opinion you can prepare for style/level of difficulty but not for sloppiness. There weren’t too many long, meandering questions this time.

Overall I’m a bit angry. I put money and a lot of time into learning this curriculum (most of which will help me at work so that’s good) and I don’t feel like this test gave candidates a fair shot at proving their knowledge.

 

[Toucan]

FRM is for FINANCIAL RISK MANAGER. There was pretty much nothing FINANCIAL in this exam.

Also, I think I could've left my calculator home and it probably would not have changed anything.

Afterwards, I felt like Operational risk and Resiliency + Current Issues were like 90% of the exam. If I had to describe this exam in 1 word, it would be : IRRELEVANT.

 

[Seelmann]

As others have pointed out, the exam felt very "off". There was a large focus on qualitative questions, most of which I could only narrow down to a 50/50 (which I then chose at random depending on if the minute on the clock was odd or even).
If someone asked me now what i think about the best way of preparing... study everything in all the books. I didnt have time read them twice as they were over 1200 pages and i only got my part 1 exam results in january + it took 1 month for the garp books to arrive... i dont know. Interested to read what you guys have to say. Good luck!

 

[wojtek]

I totally agree - it was something like FRM Part 3 Exam or FRM S (Special) Exam or .... ?

 

[ivan.jarmanovic]

I could not agree more to the above... In my opinion:

(i) clustered areas (I missed a little bit "financial risk" (i.e. market, credit, liquidity and investment risk) but can recall having an overall focus on ORR and CI);

(ii) strong focus on qualitative aspects (the few quant questions were focusing mostly on calling up formulas like the "ULC" question at the far end of the exam - questions like the one from the mock exams on the Merton PD proxy, credit exposure pre/post netting etc. which required calculation and interpretation, where unfortunately rare); and

(iii) partly confusing questions (there were a couple of questions where you were confronted with a text block of an apparent starting situation and the ultimate task was "which of the following is most appropriate to be true" were the answers where more generic i.e. without a clear link to the starting situation..)

Overall: GARP mock exams where helpful but rather disconnected from the exam which actually took place. The emphasis was not much on classic "financial risk" (market, credit etc.). The official literature had a lack of practice problems and those few were rather disconnected from the exam as well. The BT scripts, practice questions, global topic drills and mock exams were a great help in understanding the concepts (which I could already apply on the job!) and where structured in such a manner that the breadth and depth where optimal for exam preparation - clearly the exam would have been a catastrophe without BT!..

In terms of time: 4h were sufficient to get through the exam. The proctors where prepared, professional and friendly. COVID rules have been prepared and adhered in accordance with local regulations.

 

[frenchmarmot]

Suprised by the very qualitative nature of the exam. I knew there would be many text-based questions etc. but not to that degree and nature. Agree many questions were poorly written and contained vague answers at best, sometimes very difficult to choose the correct one - even on topics I knew very well - because answers were formulated in an ackward way and it was not clear which one ansswered the question best. Many times questions revolved around minor parts of the books rather than the core of the learning objectives I feel. Quite clustered on parts of the syllabus. Question style was not in line with that from GARPs mock - not talking about difficulty here, just tzpe of questions.

The few purely quantitative questions were OK. I mostly expected these kind of questions to be honest where the candidate needs to choose the correct tool or formula to answer the question, i.e. the question iteself does not point to the equation or reasoning to be used. These were fair in my opinion.

In terms of timing, it was fair. I took my time and had about 5min left at the end of the exam.

It seemed difficult and suprising to most people in my room - about 10 of us - so hoepfully scores will be scaled appropriately.

 

[ruben169]

I agree with many of the things that have been said above.

i) The exam was very light on the quant side while - given the amount of material - this would typically be the part where a decently prepared FRM part 2 candidate would lie the foundation to pass the exam. In general, I felt that many topics that appeared on the exam where quite esoteric and not related to key risk management concepts (e.g. questions on specific contractual provisions for outsourcing, indemnification provisions for fund managers).

ii) The exam had one key deficiency: GARP assigns a mountain of material to candidates but it is just not reasonable to expect that any reasonably prepared candidate will know all of them in detail. Since some of the questions (FRTB) really went into the nitty-gritty detail, preparing for this level of detail for 100 readings would double the amount of prep time with virtually no additional (job)useful return,

iii) The qualitative questions of the PE where nowhere near the actual exam and the quant. Qs where absent. As a result there is a terrible lack of suitable training material if you just buy the GARP readings. I find this really disturbing given the standards that GARP claims to set for their candidates but where they apparently do not hold themselves to.

The examination room where I sat in Amsterdam had no day light and felt a bit like a late-night club, which I personally found not contributing to my concentration either. The proctors where professional and helpful

 

[mariomansour]

Felt like about 15 out of the 100 readings were tested. Ridiculous.

Left the exam so angry, not because of the difficulty, but because it felt as if all the time & effort I put into this exam was all for nothing.

The issue is not that it was too qualitative or barely any calculations were asked, it's about the topics that they targeted. No Ho-Lee? No Vasicek? CIR? Surplus at risk? So many topics where I felt actually reflected "financial risk management" were nonexistent.

Very, very odd exam.

 

[maartenvdz]

I agree as well. Many questions felt like GARP was testing my ability to make quick judgement calls / educated guesses and select the least worst answer, if that makes sense. I am sure that some “gambles” will not have worked out in my favor so indeed the passing rate should not be set too tight. No one can be expected to have known all of the tiny details that were tested.

I was happy that they did not cancel the exam in Singapore. The Covid situation has worsened here, so I am quite sure that GARP had to pull some strings to let it continue.

does anyone know why it takes so long for GARP to release the results ? Mid July ?

 

[Rhim0001]

Agreed the exam is biased towards the qualitative side, while most of the calculation part is straightforward. I recall there are questions on Model 1, liquidation cost, repo without haircut, ULC, probability of survival on an exponential function, joint survival probability etc. Ho-Lee model, Vasicek and CIR are all missing, and there are only 1-2 on Merton/KMV, but they are not calculation questions.

As for the qualitative side, there are around 10-15 of those questions I have to choose an answer between 2 options, so I have chosen the one which sounds more reasonable to me. Just hope the cutoff score won't be too high.

 

[Rhim0001]

I agree as well. Many questions felt like GARP was testing my ability to make quick judgement calls / educated guesses and select the least worst answer, if that makes sense. I am sure that some “gambles” will not have worked out in my favor so indeed the passing rate should not be set too tight. No one can be expected to have known all of the tiny details that were tested.

I was happy that they did not cancel the exam in Singapore. The Covid situation has worsened here, so I am quite sure that GARP had to pull some strings to let it continue.

does anyone know why it takes so long for GARP to release the results ? Mid July ?

Maybe because many of the candidates have reconsidered their answers to some of the questions and changed the answer for them by rewriting their final answer next to the checkbox. These handwritten alphabets could not be scanned.

Or maybe simply they are just too lazy

 

[thanhtam92]

I agreed with everyone above. The exam was written poorly and very difficult. Even for some quantitative questions, they touched on some very specific calculations (seems less relevant) like hurdle rate, etc. like no RAROC, no TSM. I don’t know what to feel about Part II. Some questions are super long with unnecessary information that try to confuse you. Anyway, I hope I pass but I don’t have much faith.
do you know when the result will be released?

 

[jho012345]

I read results are usually released 6 weeks after the exam.

I agreed with everyone above. The exam was written poorly and very difficult. Even for some quantitative questions, they touched on some very specific calculations (seems less relevant) like hurdle rate, etc. like no RAROC, no TSM. I don’t know what to feel about Part II. Some questions are super long with unnecessary information that try to confuse you. Anyway, I hope I pass but I don’t have much faith.
do you know when the result will be released?

 

[jho012345]

I also had a similar thought during the exam about how little I was using my calculator to answer questions....to me it felt like 60%-75% of the exam was qualitative with the rest being quantitative. I took the exam in SF where there were maybe 20 folks taking the exam but with 50 seats available. I think there was a massive no-show rate for my testing site... Did others notice that as well?
Here is a breakdown of what I recall being specifically tested...note these are from the GARP study guide on changes to the FRM materials for part 2 (https://www.garp.org/media/a1Z1W000005jGggUAE) :
1. Financial Crime in Times of COVID-19 – AML and Cyber Resilience Measures”, Financial Stability Institute, May 2020. -> 2 or 3 q's
2. Global Financial Stability Report: Markets in the Time of COVID-19, International Monetary Fund (IMF), May 2020. (Chapter 1 only) -> there were 2 questions on this I think. One specifically asking what happened to IG vs. HY spreads during March 2020/April 2020....does anyone remember the second question?
3. “Cyber Risk and the U.S. Financial System: A Pre-Mortem Analysis,” Federal Reserve Bank of New York Staff Reports, June 2020. -> there were a ton of cyber risk questions. These were all framed as "best practice" type questions where the process of elimination was hard against 2 or more.
4. Valentin Haddad, Alan Moreira, and Tyler Mui, “When Selling Becomes Viral: Disruptions in Debt Markets in the COVID-19 Crisis and the Fed’s Response.” -> I don't remember this article being tested
5. Patrick Bolton, Morgan Despres, Luiz Awazu, Pereira Da Silva, Frédéric Samama, Romain Svartzman, “The Green Swan – Central Banking and Financial Stability in the Age of Climate Change,” Bank for International Settlements (BIS), January 2020. (Chapters 1-3 only) -> annoyingly long material to read and worst podcaster ever.... also NOT TESTED SMH....
6. Climate Change: Physical Risk and Equity, Global Financial Stability Report: Markets in the Time of COVID-19, International Monetary Fund (IMF), May 2020. (Chapter 5 only) -> I dont think this was tested but feel free to correct me on this
7. Stephen Cecchetti, Kim Schoenholtz, “Replacing LIBOR” https://voxeu.org/article/replacing-libor. September 2019. -> one question on Libor vs. SOFR that drove me crazy. They asked something like.... "Given all of the recent updates on LIBOR vs. SOFR, Which of the following is correct" a.) US regulators are forcing the addition of fallback language to existing libor based contracts, b.) Libor's derivatives market is greater than SOFRs c.) SOFR's derivatives market is greater than LIBORs d.) I forget.... I spent a lot of time on this stupid question as you can tell... lol

Stephen G. Dimmock and William C. Gerken: Predicting Fraud by Investment Managers (2012) -> this was definitely asked about

Quantitative areas that I recalled:
1 LogNormal Var 95% -> 1 day 99% type question
1 constant hazard rate CDF question
1 model 1 constant vol no drift two month downrate scenario calc


That's all that is coming to my mind, but hopefully others can add for additional benefit! Cheers and good luck all!

 

[jho012345]

FRM is for FINANCIAL RISK MANAGER. There was pretty much nothing FINANCIAL in this exam.

Also, I think I could've left my calculator home and it probably would not have changed anything.

Afterwards, I felt like Operational risk and Resiliency + Current Issues were like 90% of the exam. If I had to describe this exam in 1 word, it would be : IRRELEVANT.

I can certainly empathize with your stance. However because Covid forced most companies to go to a full remote working situation, BCP/DR/Ops risk is top of mind of most risk practicioners today (certainly at risk committees we're talking about it endlessly these days) and plus with markets melting upwards a lot... the market risk side seems like its the "dull" one relatively speaking! Odd times indeed and definitely odd to make a test around just the recent events... the test I think was trying too hard to be relevant and forgot to bring it back to the fundamentals (that we all crammed for) lol... just my two cents!

 

[Eustice_Langham]

Apparently those took Level 1 via CBT and finished early could find out the number of questions that they got correct via the CBT testing

 

[Ullaskarkera]

I also had a similar thought during the exam about how little I was using my calculator to answer questions....to me it felt like 60%-75% of the exam was qualitative with the rest being quantitative. I took the exam in SF where there were maybe 20 folks taking the exam but with 50 seats available. I think there was a massive no-show rate for my testing site... Did others notice that as well?
Here is a breakdown of what I recall being specifically tested...note these are from the GARP study guide on changes to the FRM materials for part 2 (https://www.garp.org/media/a1Z1W000005jGggUAE) :
1. Financial Crime in Times of COVID-19 – AML and Cyber Resilience Measures”, Financial Stability Institute, May 2020. -> 2 or 3 q's
2. Global Financial Stability Report: Markets in the Time of COVID-19, International Monetary Fund (IMF), May 2020. (Chapter 1 only) -> there were 2 questions on this I think. One specifically asking what happened to IG vs. HY spreads during March 2020/April 2020....does anyone remember the second question?
3. “Cyber Risk and the U.S. Financial System: A Pre-Mortem Analysis,” Federal Reserve Bank of New York Staff Reports, June 2020. -> there were a ton of cyber risk questions. These were all framed as "best practice" type questions where the process of elimination was hard against 2 or more.
4. Valentin Haddad, Alan Moreira, and Tyler Mui, “When Selling Becomes Viral: Disruptions in Debt Markets in the COVID-19 Crisis and the Fed’s Response.” -> I don't remember this article being tested
5. Patrick Bolton, Morgan Despres, Luiz Awazu, Pereira Da Silva, Frédéric Samama, Romain Svartzman, “The Green Swan – Central Banking and Financial Stability in the Age of Climate Change,” Bank for International Settlements (BIS), January 2020. (Chapters 1-3 only) -> annoyingly long material to read and worst podcaster ever.... also NOT TESTED SMH....
6. Climate Change: Physical Risk and Equity, Global Financial Stability Report: Markets in the Time of COVID-19, International Monetary Fund (IMF), May 2020. (Chapter 5 only) -> I dont think this was tested but feel free to correct me on this
7. Stephen Cecchetti, Kim Schoenholtz, “Replacing LIBOR” https://voxeu.org/article/replacing-libor. September 2019. -> one question on Libor vs. SOFR that drove me crazy. They asked something like.... "Given all of the recent updates on LIBOR vs. SOFR, Which of the following is correct" a.) US regulators are forcing the addition of fallback language to existing libor based contracts, b.) Libor's derivatives market is greater than SOFRs c.) SOFR's derivatives market is greater than LIBORs d.) I forget.... I spent a lot of time on this stupid question as you can tell... lol

Stephen G. Dimmock and William C. Gerken: Predicting Fraud by Investment Managers (2012) -> this was definitely asked about

Quantitative areas that I recalled:
1 LogNormal Var 95% -> 1 day 99% type question
1 constant hazard rate CDF question
1 model 1 constant vol no drift two month downrate scenario calc


That's all that is coming to my mind, but hopefully others can add for additional benefit! Cheers and good luck all!

For libor seems they had 2 questions...1.sofr libor which is more volatilie....sofr is more volatile....but if we use 3m sofr compounding rate vs 3m libor...libor is more volatlie..not sure which ans they would consider,.....2.libor still has more outstanding contracts.




Climate risk...something like mkt oper and liq risk are impacted.

Covid times....spot and future oil price fell.

Cyberrisk....effected..independent of location

UNsupervised learning...clustering

AML TF....more complex and use of ML

small retail bank...loan..3rd party...something like backtesting black hole

I recollect above 8 from current issues

 

[Rhim0001]

For libor seems they had 2 questions...1.sofr libor which is more volatilie....sofr is more volatile....but if we use 3m sofr compounding rate vs 3m libor...libor is more volatlie..not sure which ans they would consider,.....2.libor still has more outstanding contracts.




Climate risk...something like mkt oper and liq risk are impacted.

Covid times....spot and future oil price fell.

Cyberrisk....effected..independent of location

UNsupervised learning...clustering

AML TF....more complex and use of ML

small retail bank...loan..3rd party...something like backtesting black hole

I recollect above 8 from current issues

libor still has more outstanding contracts --> I recall there is an article said existing contract can still use LIBORin 2025 something like that, but maybe I am wrong
Climate risk...something like mkt oper and liq risk are impacted -> Yes, the answer is climate risk has been proven to significantly impact mkt,ops and lqd risk
Covid times....spot and future oil price fell ->Didn't read this article but I recall crude oil spot price became negative in Mar2020, so should be correct I guess?
Cyberrisk....effected..independent of location, UNsupervised learning...clustering -->Got these same as you too

There is also a question regarding the internal audit department assessing the risk in the company, what is the best practice of CRO, I was struggling with 2 options
a. The CRO should ask business unit managers to challenge their own models before submitted to approval
b. Build a contingency plan...
Option a. sounds very reasonable but I have chosen option b, since a. should be done by risk managers instead of the business unit manager themselves, but again, maybe I am wrong