Most of you knew why I chose to do the CFA. Your reasons for taking the CFA may differ, but I’d imagine we did the same thing: researching whether the quality, content and cost of the qualification fits our career goals and aspirations at that given time.
So far, so good.
But there’s one aspect that continues to confound potential CFA candidates – proper compensation data (or the lack thereof). How can we evaluate the return on investment on CFA without the pre and post qualification salary information? I mean, other qualifications such as accounting (ACCA/ACA/CIMA) or MBA seem to have at least some rough indication on this, why can’t the CFA?!
Valid questions, indeed. But there are good reasons why CFA salary data are so elusive.
Before we delve deep into the issue at hand, let’s take a step back to consider how would one properly attempt to attribute the monetary benefit of CFA.
Step #1: To do this, one needs to survey 2 large groups of people: one taking the CFA, and one not (control group). Collect salary and background data in the beginning such as location, gender, sector, job type, years of relevant work experience, number of years to pass CFA etc.
Step #2: Then, ideally, collect the same set of information from the same 2 groups annually for 5 years from today. You can already imagine how hard this step is. Imagine me filling the same CFA survey 5 years in a row? Nah.
The alternative (less perfect) method is to find, at one point in time, another 5 sets of groups of “similar” background to the first years, increasing in seniority, to form a representative group of what their salary progression might be. The benefit of this is you don’t have to wait 5 years for a set of results and you have more data over time. The downside is of course, it’s difficult to find those 5 group sets that have “similar” background itself. Bummer. It seems that CFA Institute also gave up trying this method since 2007 due to lack of “apples-to-apples” data comparison.
Step #3: Run your tests, adjusting for background data, and hope that there is a statistically significant excess compensation that the newly CFA-qualified group achieves compared to their non-CFA peers.
The steps above are a gross simplification of what actually happens in an econometrics analysis – the method of making inferences from data. I’ll spare you the details from the world of panel data regression, suffice to say that it scars you for life 😉
Now that we understand the challenges that comes with data collection for an analysis of CFA’s monetary value, can’t we at least emulate the imperfect methods of the other qualifications mentioned above?
Unfortunately, this isn’t that straightforward either! Here’s why:
A) It’s not a profession-specific qualification with clear career progression. Unlike actuaries, lawyers, accountants.
The crux of this is that CFA is not a pre-requisite even for relevant professions like equity research and asset management. In contrast to actuary /law /accounting firms, the juniors there are usually incentivised through salary increment or bonuses when they pass their professional exams. This “pass-or-out” mentality forms a clear career ladder progression and rich datasets of their associated compensation. The CIMA qualification is a great example of showing such rich salary data sets at each stage of the exam.
We don’t have that widespread incentivisation system (yet?) with CFA, as there is a bigger difference in theory vs. practice with investments, capital markets and behavioural finance. Market experience counts too. It is not as rigid as accounting studies, regulations or actuarial forecasting. You can argue that CFA is similar to MBA in this sense.
B) It’s not a full-time qualification where candidates temporarily leave full time work for. Unlike MBA or PhD.
As mentioned in the “Overview” section above, to properly attribute the increase in salary to the effect of one’s qualifications, you need to adjust for other factors. The temporary break from work to study removes the “noise” from the salary data, therefore data collection frequency is less (hence easier).
However, since CFA is mostly a study-on-the-job qualification, more frequent data collection is needed, which severely reduces the complete data sets that will be eventually collected. Do you think you would consistently fill 5 years worth of CFA survey data? Who would?
In the usual helpful spirit of 300 Hours, all hope is not lost. Here’s what you can do despite these setbacks:
Short term – Focus on Now
What should you do now regarding salary information when considering the CFA?
Just look up salary information for your target job role. That’s the best gauge for now.
E.g. if it’s equity research, take into account your relevant experience and realistically gauge what level you are suitable for.
Even if we wanted to have a shot at the ‘proper’ method, we realised it’s hard to achieve without the same access to CFA Institute’s whole population dataset.
That said, I would like to give this CFA salary conundrum a shot (we don’t give up that easily here). I’d love the opportunity to gather, analyse and share the salary analysis with all of you once it’s done!
All individual details submitted will never be shared with third parties. In other words, there’s no chance of anyone finding out your details from us.
Help me help you, by filling out the salary survey below!