Why Auditors Should be Rotated

There is a proposal afoot to mandate auditor rotation every fiver years or so.  Some don’t like it.  I think it is a great idea, with large benefits relative to the costs.

My insights, or lack thereof come from working in life insurance financial reporting in a number of different ways for around 15 years.  Only on time in 15 years, in what is arguably one of the most complex industries as accounting goes, did I ever find serious questioning going on.  Should I tell this story?

Yeah, I should, because it involves the “piece of work” that I reported to at AIG who told me, “Dealing with auditors is bloodsport.”  He also said, “Dealing with reinsurers is bloodsport.”  Delicious that this came to bite him in both ways.

A certain life reinsurer who was large then (call them Geta Life), but is out of the business now (unimaginable then, but given what happened here, no surprise), reinsured a large portion of the immediate annuities and structured settlements, including rated structured settlements that the AIG domestic life companies had written.

Did the treaties pass risk?  With a vengeance they did; not only did they pass mortality risk, but all investment risks were passed as well.  For this fine service, Geta Life earned 1% per year on the surplus relief, i.e., the difference between the book value of liabilities and assets reinsured.

It was not so well understood then, but mortality risk for structured settlements did not tend to work out well.  After an injury giving rise to a court case which would structure a settlement for the plaintiff, the defendant would ask insurance companies to bid on the settlement, which was a stream of certain and life contingent payments.  When the injury impaired the life of the plaintiff, bidding would get stiffer, because it is cheaper to fund life-contingent payments to those who aren’t likely to live so long.

Or so you would think… there was one case where a two-year old boy was injured, to the point of being in a coma, and the underwriter who bid the case rated him as having the lifespan of one who was age 73.  But the money transferred to the parents in the judgement was more than enough to care for the boy, and have a lot left over for the parents.  Ding!  The kid would live a lot longer than 15 years as a result of the settlement.

Rated settlements, where one bid on impaired lives, carried the “Winner’s Curse.” If you won, you overpaid.

But this was not on the radar screen of the somewhat oblivious Geta Life, until they found that the treaties were passing large losses to them, and they decided to audit the treaty.  Sadly, the actuaries above me, who had signed the treaties before I was employed by AIG, forgot to inform the investment department that the treaty limited the trading of around 20% of the bonds of the company in ways that would be mimicked 10 years later in CDOs.

  • Trades may not lower credit quality
  • Trades may not lower yield
  • The cashflow profile of the assets can’t be materially changed.
  • And a few more things…

This problem got dumped in my lap as a young actuary, as I found we were way out of compliance with the treaty terms, selling had gone on with abandon, on assets the reinsurer relied on, reducing the investment income the reinsurer would receive in a falling interest rate environment.

So, I proposed to the reinsurer that I go back in treaty history, and select assets purchased to replace those sold that would have kept the treaty in compliance, and put those into the segregated portfolio, and inform the investment department of the rules.  Once Geta Life understood that, they agreed to my “solution.”  That solution took me several months to work out, but I got it done.

In the meantime, the reinsurance treaties with Geta Life had become so valuable to AIG’s domestic life subsidiaries that if they came into question, the subsidiaries would fail.  The accountants of the auditors, realizing that there was something big to analyze, but not knowing how to do it, called in one of their best actuarial auditors.  My ever-confident boss knew he could beat him.

I still remember critical parts of the meeting as the actuarial auditor slowly checkmated my boss, and forced him to reduce the reserve credit for GAAP accounting, resulting in a sizable loss.

Closing off the story, Geta Life was satisfied on the changes in the asset portfolios, but was still annoyed at the losses.  The changes in assets did not avail much; bad underwriting was pinching.  They came to us saying that “Reinsurance is a good faith venture. You’re not supposed to take advantage of us.  Refund our losses, or we will take you to court for not having having managed the treaty properly from inception.”

I said, “You wrote the treaty.  You accepted my asset changes.  You are supposed to absorb mortality risk, such as there is.  I am not an officer of the company, so if you aren’t happy with this, talk to my boss.”

Shortly after that, I left AIG; it was not a great place to work.  Geta Life sued AIG for damages and won (far more than they should have).  Should have produced a blip in terms of earnings, and didn’t.

Mmmmm…. back to the original point.  Should auditors be rotated?

In all my years of financial reporting, I got wind of things in other areas of the company that I served that auditors should have questioned.  Auditors have often been “lap dogs.”  Only once did I ever see a significant challenge.  More often, I saw the auditors try to help the company explain an “accounting oddity.”  (AIG had a nonstandard way of reporting deferred annuity reserves that was very liberal, and it was proposed by their auditors.)

If auditors know that they are only going to be on the job for five years, they will realize a few things:

  • If this is going to die in a few years, it doesn’t matter as much if it dies next year.  Maybe firm reputation is worth more than two more years of a contract.
  • My work will be reviewed by someone unsympathetic to me in a few years.  He will have little incentive not to tear my work up, and call for restatements.
  • Having a fresh set of eyes on corporate finances will lead to questioning of assumptions that get ignored because they are boilerplate to the continuing auditor.
  • If auditing ceases to be an annuity to auditors, they will be less complacent, and might even act like auditors on occasion.

My experience with auditors was that they spent a lot of time on the data, and rarely asked the tough questions on assumptions and methods.  They were bean-counters, not actuaries, and certainly not businessmen.

If auditors are rotated, the incentives for just letting things slide will diminish.  That’s why auditors should be rotated.


  • Conscience of a Conservative says:

    I’ll be more banal on the subject. Sarbanes Oxley recommends auditor rotation(I believe every five years). Auditors on the job for long periods of time grow complacent. In addition the auditing firm should not be advisory and consulting services beyond the audit. I believe the only reason firms that receive auditing services fight this common sense recommendation is the knowledge that rotating auditors can be taxing on resources, especially on the CFO in particular.

  • Greg says:

    >> “If auditors are rotated, the incentives for just letting things slide will diminish. That’s why auditors should be rotated.”

    Sounds like a great way for each of the auditors to blame the others / no one to accept responsibility for anything

    Maybe you think the “new” auditors would want to clean all the skeletons out of the closet — but then the “old” auditors would want to do the same somewhere else.

    The incentives would be for all auditing firms to sign off on each other’s decisions, collect their fees and hit the golf course early.

  • bungaloo says:

    David…many years ago and it may still be the case, the big 5 Canadian banks had to have two auditors. One was “permanent” and the other would rotate every 2-3 years (I forget the actual number). As an accountant (not an auditor) I can say there is a lot to be said about having continuity on the audit team. Many a newbie auditor (and not just the junior auditors) has left a meeting I have been in dumbfounded because they do not understand the business and the specific intricacies of the business as it exists in the company. Keeping a permanent auditor allows this continuity to be maintained. Having a second rotating auditor gets a fresh set of eyes on the books regularly. While this is expensive and probably best implemented by the 500 biggest listed companies, I think it would be a far less expensive, less onerous and more protective than thousands of pages of SarbOx documentation that only gets reviewed when internal audit comes in.

  • Conscience of a Conservative says:

    David I think there’s a bigger issue. Audits are done for the stock holders and bond holders benefit, but somehow that fact has been subverted and the auditors and regulators now believe the audit is done for the managers and regulators first and foremost.

  • Pacioli says:

    The blog author’s ‘insights’ do not support the four concluding points at which he arrives. The author has mentioned elsewhere that he has never taken a course in accounting, much less been an auditor. This is not a knock, for the author is clearly well-versed in other areas. But let’s take the four points in turn:
    1) “If this is going to die in a few years…”
    While it is indeed tempting to think that mandatory rotation would encourage quicker, more pro-active challenging of management’s assertions and methodologies, in the real-world environment of maximizing effectiveness/efficiency in a limited amount of time on the audit, the actual result is that the audit teams will conclude “it will not be our responsibility in a few years”.
    2) “My work will be reviewed by someone…”
    This already occurs in the form of PCAOB-mandated firm peer reviews of each other’s audit engagements. So this point is null in supporting mandatory rotation.
    3) “Having a fresh set of eyes…to question assumptions…”
    Firms already rotate the audit partner internally every few years. This leads to the desired outcome of having fresh eyes periodically evaluating assumptions, while not losing all of the (enormous) intellectual capital that has been built up by the rest of the audit team over time as they have learned the specific business and general industry of the respective client.
    4) “If auditing ceases to be an annuity…”
    Again, mandatory rotation would do nothing to address this (supposing, for a moment, that the point is even valid). If anything, audit fees would increase because the firms ‘rotating in’ would always be charging higher audit fees than before, due to the greater resources required to perform first-year audits.

    @CofC –
    You are indeed correct that “there’s a bigger issue”. But you are incorrect in your identification of it. Who pays the audit fees? Who funds the regulator (the PCAOB)? Once you answer those questions, you will have your answer to what the ‘bigger issue’ is.

    • Conscience of a Conservative says:

      Management may pay the auditing fee, but it is for the the stock holders and lenders. To me this is a legal issue. Courts need to empower stock and bond holders to sue in court when auditors do not release important info or fail to do their job. The SEC has been known to keep their hands on info learned from audits that flagged bad practices but weren’t released.

      • Pacioli says:

        @ CofC –

        Incorrect. Management does not pay the audit fee. The Board does.

        Regardless, I’d be interested how you support the notion that the audit is “for the the stock holders and lenders” when they are not the ones hiring/firing/paying the auditor. Very curious how you arrive at that assertion.

        “Courts need to empower…when auditors…fail to do their job.”

        I’d also be curious as to how you define the auditors’ job. In the general public, there is an enormous gap between what people perceive the auditors’ job to be versus what it actually is (if you read the audit report).

        Again just curious as to whether you are operating under realistic definitions here.

        • Conscience of a Conservative says:

          The board is there to represent the stock holders. Companies engage external auditors to assure stock-holders the numbers are legit. It’s that simple.

    • Pacioli, CoC, fair points. My biggest problem with audits while on the receiving end as an actuary was how poor they were. They focused on things that didn’t matter, and the relationships between the firm and auditor seemed to implement control over the auditor, and reinforce audits that never asked the deeper questions.

      Sometimes I wonder if we would be better off eliminating audits to eliminate the sham that they sometimes are. I have suggested previously that having the audit committee of board get a real budget, and hire the external auditors. That doesn’t entirely eliminate the agency problem, but it would place the responsibility in the hands of those who are supposed to watch out for the shareholders.

      Thanks for commenting.

      • Pacioli says:

        “I have suggested previously…”

        Your suggesting is already reality. That is, the hiring/firing/payment of the auditor is already under the purview of the Board. So that dynamic is already as it should be.

        “…wonder if we would be better off eliminating audits…”

        I fail to see the benefits of this outweighing what would certainly be enormous costs.

        @ CofC –

        “Companies engage external auditors to assure stock-holders the numbers are legit. It’s that simple.”

        In general, it is that simple. But in specific, what level of materiality threshold is embedded in you standard of “legit”?

        That is the essence of my question, and one of the main drivers of the enormous gap between reality and most of the public’s perception.

        • Pacioli, fine. Are you saying that there isn’t anything anything wrong with the auditing process at present?

          Positively, what would you recommend to improve audits then, because have been of the other side of them, they don’t engender much confidence for me.

          • Pacioli says:

            I am saying I have not seen any specific proposals for improvements where the benefits outweigh the costs.

            Every process in the world has something wrong with it. This one is no different.

            But the proposed auditor rotation addresses none of the current shortcomings. Importantly, it would impose incremental costs.

          • Pacioli — two notes from Tom Snelling:



            I’m not convinced that there is nothing to be done to improve auditing. Research1234 has a good idea for insurance auditing: train auditors in broad on actuarial topics. Perhaps the same could be applied to other industries and their nuances. That said, life insurance accounting is among the most complex of all industries, maybe there is not so much to learn elsewhere.

            Thanks for commenting. My opinion is now open here.

          • Pacioli says:

            Thank you for including the articles from Snelling’s blog. Some interesting insights there.

            Even Snelling acknowledges that auditor rotation is nowhere near the most important thing to address. No, it is the FASB (quality of the standards) and the SEC (the oversight body is stretched too thin – wears too many ‘hats’ and lacks competence in most of these areas).

            Regarding your endorsement of Research1234’s proposal – that should also already be the case. If you were not receiving that level of client service from your auditor, then you (not you individually, your company) should have demanded it.

            In my experience as a Big 4 auditor, audits of specialized industries (like insurance, or asset managers, in my case) always included members of the audit team (usually higher level) that had relevant experience and in some cases, had even attained industry designations (such as CFA, in my specific case).

            So, yes, the audit team should possess the subject matter competence to complete an effective audit. Ultimately they are client service organizations, and if the client does not demand this appropriate level of service (or take their business elsewhere), then it seems to me that some of the blame falls there, too.

          • Pacioli says:

            Some decent insights around what an audit is, and what it is not:


        • bungaloo says:


          I am quite sure there are many things that can be done to improve auditing. As for specific training, I currently work in the reinsurance division of a large insurer and our audit team has an actuary on board. I don’t know how much “auditing” he does but he is a key part of the audit team. Maybe there is a middle ground where auditing firms use some subject matter experts trained in auditing techniques and some auditors trained in the specific fields they are auditing. So the CPA gets some actuarial training and an actuary gets some audit training and the two are an audit team.

          One thing that has always driven me crazy about many auditors is their lack of non audit experience. I cannot tell you how many auditors I have had to deal with that just do not have the foggiest idea how business really works (and surprisingly quite a few who have far less of a solid grasp on accounting concepts but that is an entirely other rant)

  • Conscience of a Conservative says:

    Francine Mckenna has a great blog on the subject. Re-The auditors

  • Conscience of a Conservative says:

    I don’t think the solution is all that hard. Rotating auditors is a good step, as well as increasing the Pay and responsibility of the board of directors. To make it all work though the SEC needs to increase the ability of stock-holders to now hold those people accountable through share-holder proxies and the ability to use the courts.

  • Research1234 says:

    David, my in my experience as a consulting pension actuary I found very few auditors understood actuarial principles well enough to really challenge the assumptions or methods, let alone the results. Sounds like you had similar experiences.

    I think that more in depth actuarial training for auditors will solve the problem you identified.

1 Trackback