Why CMA, CFA, and CPA Get Called Mandatory, And What That Word Is Actually Hiding

2026-09-14

The posting says CMA preferred. Not required, not mandatory, preferred, in the soft language job postings use when they want to sound serious without quite committing to anything. I have read enough of these to stop believing the word means what it says. Below it, the same posting also lists CA and MBA as acceptable. Three different three-to-seven-year investments, offered as though they were interchangeable line items on a checklist, and somewhere underneath that casualness sits a question nobody applying for the job ever quite gets a straight answer to: do I actually need this, or does the posting just want to look serious?

I went looking for a straight answer and found, almost immediately, that the word CMA does not refer to one thing. There is an Indian CMA, run by a body called ICMAI, created by an act of Parliament, and there is a US CMA, run by a private American membership organization called the IMA. They share three letters and almost nothing else. The Indian one carries a specific legal power, it is the only credential in the country legally allowed to sign a statutory cost audit for certain categories of companies under the Companies Act. The American one carries no such power in India at all; it cannot sign anything a court would recognize as official. The job posting that said CMA preferred gave no indication which of the two it meant. Multinational and GCC postings, it turns out, almost always mean the American one. Indian manufacturing postings asking for cost-audit work almost always mean the other. Nobody tells you this. You are expected to already know, or to guess, or to find out only after you have registered for the wrong one.

Even once you have sorted out which CMA you are chasing, the deeper question is whether any of this actually gates the job, or whether preferred really does mean preferred. The evidence, once I went looking, pointed in an uncomfortable direction: not really, and also, it depends which door you are trying to open. At the hiring stage, credentials behave less like a locked gate and more like currency that can be substituted for other currency. A CA candidate who describes their own work as "assisted" has been shown to rank below a CMA candidate describing the identical kind of work as "owned", in that specific pairing, a sentence beat the letters after the name. Analyst-level roles at GCCs regularly accept an incomplete CA or a plain MBA as sufficient. The full credential, any of the three, is often not the actual bar being cleared.

Promotion looked like it might be a different animal, an internal leveling framework, the kind of document that says a Senior Analyst needs to be pursuing or holding a specific credential before the title changes, enforced by HR rather than negotiated with a hiring manager. I went looking for one of these documents and could not find it; companies do not publish their internal promotion criteria for a stranger with a search engine to read. What I could confirm, indirectly, is that the credential bar visible in public job postings does harden as the seniority listed goes up, an incomplete CA or an MBA is often enough at analyst level, a completed CA or a CPA starts to matter more for senior roles, which is consistent with promotion working as a harder, more procedural gate than hiring, without quite proving it.

And here the investigation ran into something that changed how much I trusted anything I had read up to that point. Almost every number claiming a certification raises your salary, every headline compensation percentage, every confident lakh-per-annum figure, traces back to a small circle of sources with an obvious stake in the answer: the coaching institutes selling exam prep, or the certifying bodies themselves. I checked the actual primary source behind one of the most commonly repeated average-salary figures for the Indian CMA. It was calculated from two anonymous salary submissions on Glassdoor. Two. That number has since been copied across dozens of coaching blogs as though it were a robust industry survey, and at least one of those sites was honest enough to admit its own figure ran higher than the older estimates still floating around on other coaching pages, which is its own kind of confession, since it means none of these sites are actually measuring the same thing. They are each just repeating whatever number sounded credible the last time someone asked.

If the salary claims could not be trusted, the demand claims at least had numbers attached that were harder to fake, actual candidate counts, actual member counts, published by the credentialing bodies themselves rather than a coaching site repeating a coaching site. And those numbers told a story with a shape to it, not a flat yes or no. Roughly fifty thousand candidates are expected to sit CFA exams in India this year, with new registrations growing twenty percent annually for three years running. The number of people who actually hold the full charter in India, after all three levels and the required years of work experience, is a little over five thousand. The same shape shows up on the Indian CMA side: seven hundred and fifty thousand people registered as students, only a hundred thousand who have actually qualified. Whatever inflation is happening in this market is not happening evenly. It is concentrated almost entirely at the entry of the funnel, the pursuing, the partially cleared, the "Level 1" line on a resume, while the far end of that same funnel, the fully qualified population, stays comparatively small, protected less by anyone's virtue than by a brutal, multi-year attrition rate that has nothing to do with how many people start.

That protection, though, turns out to be a policy choice rather than a law of nature, and policies move. The US CPA license has historically required a hundred and fifty credit hours of education, more than a standard three-year Indian B.Com provides, which is exactly why Indian aspirants have spent years bridging the gap with a master's degree or a partial CA before they can even sit the exam. That barrier used to be close to absolute. In 2025, the American licensing bodies approved an alternative path, a hundred and twenty credits plus two years of supervised experience, instead of a hundred and fifty credits, and more than twenty states adopted it within a year. The exact wall that was keeping the fully licensed CPA population small is being taken down, piece by piece, while I was in the middle of writing this. Whatever scarcity the completed credential carries today is not guaranteed to still be there in five years, and there is no particular reason to assume the other two certifications are exempt from the same kind of quiet erosion, only that I have not yet found evidence either way.

Which raised the harder question underneath all of this: if it is not really testing content, exactly, and it is not really a legal requirement outside one narrow case, what is a multi-year credential actually filtering for? The syllabus itself is mostly not the accounting most people assume it is, the CMA leans heavily toward budgeting, forecasting, and decision analysis, the language of FP&A rather than the language of closing a ledger. But content was never really the bottleneck. The real filter sits somewhere quieter: nearly nine out of ten people who start the CFA program never finish it, and the dropout is heaviest in the middle stretch, after the first level's adrenaline has worn off and before the finish line is close enough to pull anyone forward. Nobody is checking on you during that stretch. No professor takes attendance. No manager assigns you a deliverable. There is a natural, quiet off-ramp available every single month, and most people eventually take it, not because they failed anything, but because nobody was there to notice if they simply stopped. The CFA Institute itself seems to have concluded the same thing: starting in 2027, it is eliminating the option to pay and defer an exam sitting, specifically because deferring correlates with losing momentum and never coming back. An institution does not close a door like that unless it has watched enough people quietly walk through it.

I wanted this to be a clean story about grit, about a credential rewarding the specific, rare trait of sustaining yourself without anyone watching. It mostly is, but not as cleanly as I wanted. I went looking for a comparison case: a similarly difficult Indian credential that does come with real external structure, to see whether the structure changed the pattern. Chartered Accountancy fit, a mandatory three-year articleship under a supervising CA, real attendance, a real stipend, real oversight. And CA's own numbers offer little comfort for the structure theory: its Final-level pass rate per sitting sits around eleven to fourteen percent, comparably brutal to CFA's per-level odds, despite three years of supervised training sitting underneath it. The reason, once I looked closer, is almost embarrassing: the articleship supervises the workplace training, not the studying. The actual papers still have to be prepared for alone, on your own time, with the same absence of anyone checking in that the CFA has. So the comparison never isolated what I wanted it to isolate. What I am left with is a weaker, more honest claim than the one I started with: wherever the studying itself is unsupervised, attrition concentrates there, regardless of what surrounds it, which is consistent with self-direction being the real filter, but does not rule out that the people with more room in their lives, financially or otherwise, are simply the ones with enough slack to survive that stretch. I could not separate the two with what was publicly available, and I am not going to pretend I did.

All of which finally gets to the question that actually matters to someone deciding whether to start: is it worth it, and does the answer change depending on where you are standing when you ask it. The costs themselves barely move with career stage, a US CMA runs roughly a lakh and a half to two lakh all in, a CPA closer to two and a half to four and a half lakh once the credit-hour bridging a three-year Indian degree usually needs is accounted for, a CFA charter somewhere between four and eight lakh across all three levels depending on coaching and retakes. What changes is not the rupee figure. It is which resource you are actually spending.

A fresher paying that cost is mostly paying in time, years of the specific, narrow window when a person has the fewest competing claims on their evenings, spent on a credential whose payoff, going by everything above, mostly shows up as being one of several acceptable ways to clear an entry-level screen, not as a guaranteed premium. A professional six or seven years in, chasing a named grade on an internal ladder rather than a first foot in the door, is usually spending less discretionary time and more borrowed time, evenings stolen from an already full job, sometimes offset by an employer willing to foot the bill in exchange for a retention bond, which freshers are rarely in a position to be offered in the first place. Neither version is obviously the better trade. They are simply different currencies, and which one a person happens to have more of, right now, probably matters more to the honest answer than anything printed on the certificate itself.

Even that framing, fresher against six-years-in, turned out to be a cleaner split than most people's actual lives allow. The credential's own multi-year timeline runs straight through the years when a person is neither purely a fresher nor purely established. Most people actually going through this are starting it near the beginning of their careers and finishing it somewhere in the middle, which means they are not really choosing between the fresher's calculation and the six-years-in calculation. They are living through both, in sequence, inside a single unfinished exam.

I started this wanting an answer to a fairly small question, the one sitting at the top of that job posting: is the certificate mandatory. What I found instead is that the word was never built to survive being asked seriously. Mandatory for which version of the credential, tested against which mechanism, at which end of the funnel, filtering for which trait, paid for out of which currency, the posting was never going to specify any of that, and neither was the coaching page offering a confident lakh-per-annum answer instead, because uncertainty has never sold a course. The honest version of the answer is the one no landing page carries: it depends on a version of you that the question, as usually asked, never bothers to name. The certificate, once earned, will not have changed that. It will just be sitting there, three letters after a name, waiting to find out which version of you actually showed up to collect it.

Notes: The distinction between the Indian CMA (ICMAI, a statutory body under the Cost and Works Accountants Act, with sign-off authority on statutory cost audits under the Companies Act) and the US CMA (IMA, a private membership body with no legal standing in India) is drawn from ICMAI's own published mandate and general industry reporting on the two credentials. Resume-screening behavior and analyst-level qualification substitutability are drawn from current GCC hiring and career-guide reporting. The Glassdoor salary-sample figure, and the broader pattern of coaching-site salary claims tracing back to thin or self-interested sources, reflect a direct check of the primary listing itself rather than a secondary summary of it. CFA candidate and charterholder figures for India, and the CFA Institute's 2027 deferral-elimination policy, are drawn from CFA Institute's own reporting. ICMAI student and member totals are drawn from the institute's published figures. The CFA program's overall completion rate and its concentration of dropout around Level II are drawn from CFA Institute exam data as reported in independent exam-prep analysis. CA India's per-sitting pass rates are drawn from ICAI's published results. CPA credit-hour requirements and the 2025 alternative licensure pathway are drawn from AICPA/NASBA reporting relayed through US-CPA education guides for Indian candidates; state-level adoption is described as of 2026 and may have moved further since. Certification cost ranges for CMA, CFA, and CPA are drawn from several current Indian coaching and exam-prep sources and should be read as indicative bands rather than fixed figures, since they vary by provider, number of retakes, and the exchange rate on the day of payment.