It’s a lazy weekend, so what better thing to do than delve into ANAB’s published data on the suspensions and withdrawals? ANAB publishes the data on its website (here and here), but assuming you don’t want to poke around in it, I’ve done the work for you.
Suspension of Disbelief
“Suspensions” are the temporary placement of a CB’s accreditation “on hold” while it fixes whatever particular problem ANAB identified. An analysis of suspensions shows that between the years of 2001 – 2016, there have been only 157 suspensions, and many of these were “bulk” suspensions. ( A bulk suspension is where a CB is accredited to audit for multiple standards — say ISO 9001, ISO 14001 and AS9100 — but then has a single problem with ANAB that results in them being suspended for all their standards simultaneously. This artificially inflates the numbers since it’s technically a single suspension that just impacts on multiple certification schemes. Simply paying the ANAB bill late, for example, could result in an entry of 10 suspensions if the CB has 10 accreditations with ANAB.) If we remove the artificial inflating of bulk suspensions, we find that ANAB has only issued about 58 suspensions in the past 15 years; of those, 13 were voluntary suspensions initiated not by ANAB, but by the CB itself.
Given what we know about CB performance and behavior, that’s a stunningly low number. We are supposed to believe that over the course of 15 years, ANAB found only 33 reasons to suspend CBs for violations of ISO 17021-1. Given that I can find 18 violations during a single audit with a single CB — all while the ANAB auditor sleeps during the witness audit — this is abdication of duty on a grand scale.
Looking at the data, registrar Perry Johnson Registrars (PJR) has had the most suspensions, followed by ABS-QE. Keep in mind, this is also somewhat due to an inflation due to bulk suspensions, but it is also particularly interesting since PJR has had a troubled history with ANAB over its lifetime. When ANAB was originally just the Registrar Accreditation Board (RAB), RAB pressured PJR to split its consulting and registrar businesses. Later, RAB was split for much of the same reason, forcing ANAB to take over accreditation and handing over training to RABQSA (now Exemplar). Now it’s gone full circle, with ANAB back in the consulting and training saddle for its lab accreditations, which must drive PJR crazy. I doubt there’s any love lost there, and it may be why ANAB has a heavy hand with PJR and not so much with other CBs.
Next we look at ANAB suspensions by year, and we see a spike in 2011, and then a fairly steady dropoff thereafter. In that year, ABS-QE and NSF-ISR underwent bulk suspensions, so it’s artificial. In both cases, the suspensions were also lifted very quickly.
What caused the suspensions? According to the data, the number one reason is “nonconforming,” which technically means the CB faced an ISO 17021-1 nonconformity and didn’t respond in time, so their accreditation was temporarily placed on hold until they fixed the problem. This is fairly innocuous, since sometimes CBs can’t implement a fix within the time frame set by ANAB, or they contest the finding and ANAB keeps the clock running anyway.
The second highest source of suspensions is, to no one’s surprise, “financial” — meaning the CB simply didn’t pay their ANAB bill. My gut tells me there’s some corruption in the data here, as I’d expect to see financial have a bigger slice, so I wonder how many suspensions listed as “nonconforming” were ultimately due to a financial problem. But it’s just suspicion, and the data says otherwise.
What’s interesting to note is that a large portion of suspensions are “voluntary,” meaning ANAB didn’t’ actually do anything, and the CB just took a temporary suspension as part of normal business. This is common when a CB is upgrading to a new standard, or performing some other internal change, and they don’t expect to meet ANAB deadlines.
So how long do these suspensions last? The historical average seems to be about 64 days, or two months.
The Withdrawal Method
The ultimate penalty by ANAB is “withdrawal,” which is where the CB loses its accreditation entirely. Withdrawal can occur if ANAB cuts the CB loose due to a failure to correct a nonconformity, or it can occur if the CB quits ANAB and contracts with a different accreditation body entirely, such as UKAS. There’s a lot of “accreditation body shopping” that goes on, and we see — for example — BSI bouncing back between them regularly, presumably to get price breaks or in an attempt to find the accreditation body that will give them the least amount of oversight.
Again, a single CB can been withdrawn for their entire suite of certification schemes, artificially inflating the numbers. If we look at withdrawals by registrar, we see that AQA faced the most withdrawals, but digging into the data finds that it was voluntary and they may have just gone out of business entirely.
That last chart isn’t very useful, so let’s look instead at the reasons that CBs lose their accreditation. This chart is the most damning for ANAB, in that it shows that 92% of the time, the CB withdrawal is not the result of ANAB oversight, but because the CB either pulled out of ANAB on its own, or the particular standard was withdrawn so the withdrawal was automatic. In only 8% of cases did ANAB yank a CB’s accreditation as part of nonconformance oversight.
Who were those 8%? Nearly no one you’ve ever heard of, and not a single major CB player. AGS lost its accreditation in 2002 and 2003, and then found itself in court years later for continuing to use the logos anyway; AGS went on to become the US’ largest certificate mill, “accredited” by a company owned by the same guy who runs AGS. The others are a list of unknown registrars that have since either gone out of business or were bought by larger, better-known CBs: BSA, CALMECAC, CQR, EQAM, Excalibur, ICL, LLD, Northstar, QSI and TP.
What you notice missing from that list are the CBs that provide the bulk of ANAB’s revenue, such as SGS, NSF, DNV, Intertek, NQA, etc. What we are to believe is that these larger registrars are bulletproof, and that ANAB has never once encountered a situation grave enough to punish them with a withdrawal, even as it beat up lesser-known CB shops for lesser infractions. Given that we’ve very clearly exposed the ISO 17021-1 violations by major ANAB-accredited CBs right on this website, this can’t be accidental.
The data is weak, but even still it points to a glaring truth: ANAB is not really interested in doing it’s one job: enforcing accreditation rules to ensure the validity and trust in resulting ISO management system certifications. Instead, it’s there to collect money and smooth over the ripples when bad actors do bad things. They occasionally throw some poor smaller schmuck under the bus as a token, but would never — never — take on a big player, no matter how egregious the violations.
Now you’ve got some shiny Excel charts to back it up. You’re welcome![UPDATE 12 December 2016: A quick review of the Dutch Accreditation Body RvA’s suspensions and withdrawals page shows pretty much the same trends: smaller, unknown CBs are withdrawn permanently, while larger ones are only temporarily suspended for reasons unknown. Right now LRQA is under a curious suspension for all of its RvA accreditations, which makes me think they didn’t pay their bill. But I suspect there’s no chance at all that RvA would ever permanently withdraw LRQA’s accreditation, no matter how many dirty tricks they get up to.]
About Christopher Paris
Christopher Paris is the founder and VP Operations of Oxebridge. He has over 30 years' experience implementing ISO 9001 and AS9100 systems, and is a vocal advocate for the development and use of standards from the point of view of actual users. He is the author of Surviving ISO 9001 and Surviving AS9100. He reviews wines for the irreverent wine blog, Winepisser.