Could You Have Predicted Diamond Foods Accounting Fraud?


Diamond Foods (DMND) is being investigated by the SEC and DoJ for potential accounting fraud. I will go through a few methods to detect earnings quality, manipulation and aggressive accounting to see how well it comes up against DMND. So as we go through this, the question is, could you have identified DMND’s accounting shenanigans in advance by using such methods?


Written by

Jae Jun

follow me on

Facebook

Twitter

Here is a good case study to go through.

Diamond Foods (DMND) is being investigated by the SEC and DoJ for potential accounting fraud.

I will go through a few methods to detect earnings quality, manipulation and aggressive accounting or accounting fraud to see how well it comes up against DMND.

So as we go through this, the question is, could you have identified DMND’s accounting shenanigans in advance by using such methods?

Accounting Fraud Accusation against DMND

From the Forbes article,

“Allegedly, the company delivered so-called “momentum payments” to walnut farmers in order cook their earnings filings.”

Momentum payments mean that DMND paid their suppliers in advance. Future expenses is shifted to an earlier period which will inflate earnings.

If the audit proves to be true, then based on a Bloomberg magazine article (no link sorry), DMND’s 2011 earnings will be reinstated from $2.22 to $1.14. That’s close to 50%. Ouch!

Earnings Shenanigans

In Financial Shenanigans, shifting expenses to another period is clearly defined as one of the shenanigans used to manipulate earnings in accounting fraud cases. Now in the Bloomberg article, Mendes, the CEO, is portrayed as a very ambitious and business aggressive man. If the article is true, that “could” explain the aggressiveness in the accounting.

If you have read The Art of Short Selling, you will know that there have been many companies throughout history where such CEO characteristics have led to accounting fraud.

Balance Sheet Check

Let’s get down to the numbers.

Take a quick glance at the balance sheet. Even though a company like Diamond Foods is all about the brand of its products, the balance sheet is very unattractive. Just looking at the total intangibles should immediately raise red flags.

66% of total assets is made up of intangibles.

accounting fraud

By the numbers – accounting fraud

Accrual Ratios

Previously I went through examples of how to analyze accruals. One of the examples featured Dolby (DLB) where accruals were consistently above 25% for 4 years, yet DLB increased cash and reduced debt. DLB’s Sloan ratio was higher than the recommended 8%, but nothing jumped out and the result was inconclusive.

However, take a look at DMND below. (These numbers are from my stock analyzer and the model will be included in the next update.)

Accounting Fraud by the numbers

Cash balance is close to zero, debt has sky rocketed with all the acquisitions, net income is increasing but cash flow from operations is erratic. Dangerous signs already.

The accruals for 2009 and 2010 is shocking. In 2010 and 2011, there is a very good chance that DMND’s growth in EPS came from accruals. In other words, low quality and cookie jar type earnings.

Beneish M (Manipulation) Score

The third method you can use is the Beneish M score to detect earnings manipulation. Full details of how this model works is in the Beneish M Score article.

A score greater than -2.22 indicates a strong likelihood of a firm being a manipulator.

Coincidentally, the 2009 and 2010 numbers are red flagged.

Conclusion?

By checking the balance sheet, accruals and M score, warnings are flashing everywhere. DMND is innocent until proven guilty, but by my definition, shifting expenses to inflate earnings is fraud.

Summing up, do you:

Disclosure

No position in DMND

  • Pacioli

    “Momentum payments mean that DMND paid their suppliers in advance. Future expenses is shifted to an earlier period which will inflate earnings.”

    What??? Shifting future expenses into the current period will REDUCE earnings, not inflate them. Think about it.

    “If the audit proves to be true, then based on a Bloomberg magazine article (no link sorry), DMND’s 2011 earnings will be reinstated from $2.22 to $1.14. That’s close to 50%. Ouch!”

    I highly doubt this – especially given the absence of any link.

    “…there have been many companies throughout history where such CEO characteristics have led to accounting frauds.”

    Unfortunately, it is not that easy, because there have been exponentially more CEO’s with the same characteristics who run great companies that use proper accounting.

    “Cash balance is close to zero, debt has sky rocketed with all the acquisitions, net income is increasing but cash flow from operations is erratic. Dangerous signs already. The accruals for 2009 and 2010 is shocking. In 2010 and 2011, there is a very good chance that DMND’s growth in EPS came from accruals. In other words, low quality and cookie jar type earnings.”

    First, how is “erratic” cash flow a “danger sign”?? Just because cash flow does not go in a smooth trend does not mean that it has been manipulated in any way. For just as you stated earlier in the post, DMND makes momentum payments to suppliers. Thus, by definition, working capital is being paid in advance of the related activity. So cash flow, by necessity, is not going to be matched up by period. One could argue the business sense in this strategy. But you’ve presented no evidence to suggest any irregularity in the accounting.

    Secondly, please explain how a large percentage of EPS coming from accruals automatically equates to “low quality and cookie jar type earnings”. (and please do not point to the links you already provided – those are not compelling in the least). Accruals are simply the result of matching earned income with the related expenses incurred. Just because GAAP earnings may not fall in the same period as the cash changes hands (due to the working capital strategy we just outlined) in no way suggests that the matching principle has been violated. Better evidence must be demonstrated to try to make that argument.

  • Pacioli

    Why does it take so long for comments to appear? If you’ve got the spam protector, do you still have to comb through each one meticulously? I think it helps with reader engagement if the comments are posted timely.

  • @ Pacioli,
    Your comment was marked as spam because it was too long.

  • @ Pacioli,

    What I meant was, future expenses shifted to an early period will deflate current earnings but inflate future earnings. If they shifted 2011 expenses back in 2010, then the 2011 will be inflated.

    As for the bloomberg article, get the latest bloomberg magazine and you’ll see it in there.

    Try to keep everything into context instead of analyzing it piece by piece which won’t help. An aggressive CEO running a company based on big acquisitions, huge intangibles, trying to buy Pringles with its stock, growth in net income not matching cash from operations, increasing debt and huge accruals.

    How does this NOT show up as warning signs?

    (I recommend the research papers on accruals and how it can be used as a cookie jar.)

  • Pacioli

    “How does this NOT show up as warning signs?”

    You mention “…an aggressive CEO running a company based on big acquisitions, huge intangibles, trying to buy Pringles with its stock…” in your comment. These could all be very legitimate warning signs.

    But my questions (the last 3 paragraphs of my first comment) relate entirely to your contention that the mere existence of large accruals constitutes “dangerous signs” and represent “low quality and cookie jar type earnings.” But there is a yawning gap in your arguments. Just because a company has accruals does not mean their methodology is wrong. You must present evidence that their methodology for computing the accrual is flawed. To use your ‘cookie jar’ expression, the only way that can happen is if a company has made an erroneous/wrong accrual in one period that later reverses to affect a later period’s results. But if the accrual methodology is sound, then there has been no violation of the matching principle. And in that case, the existence of accruals (even if they are large) actually makes the financials more accurate, not misleading.

    So again, if you have no evidence that the accrual methodology is faulty, then it makes no sense to argue that the mere existence of accruals are a “warning sign”. How many (good) companies do you know of that operate completely on the cash basis? If they don’t, then they must have accruals.

  • Which again takes this back full circle.
    I know you are saying that looking at accruals does not indicate fraud. I agree because I even wrote “One of the examples featured Dolby (DLB) where accruals were consistently above 25% for 4 years” and the result was “inconclusive”.
    But what I am saying is that looking at accruals along with other manipulation detect techniques will help you identify fraud. Note that the topic of this article is not “DMND Accruals Show Fraud” and is why I ask that you read in context.
    I even linked up the argument by comparing the 2009-2010 numbers with the M score and concluded with the statement “By checking the balance sheet, accruals and M score, warnings are flashing everywhere”.

    So to keep this actually constructive, since it seems like we are both talking about two different things, I’d like to see an example of a company with huge accruals over several periods (80%+) to see how well it performed over time.

  • Pacioli

    “Note that the topic of this article is not “DMND Accruals Show Fraud” and is why I ask that you read in context.”

    The title is “Could You Have Predicted Diamond Foods Accounting Fraud?” And then the post talks about intangibles and accruals. So that then implies that eaither “DMND Accruals Show Fraud” or “DMND Intangibles Show Fraud”, or some combinatino of both. Either way, in the title you are not even acknowledging the possibility there is NO ACCOUNTING FRAUD.

    “I’d like to see an example of a company with huge accruals over several periods (80%+) to see how well it performed over time.”

    Now this could be a constructive exercise. If you could please define exactly what you mean by 80%+ (what divided by what = 80%+), then I will do some research.

    Thanks.

  • walnutexpert

    “Momentum payments mean that DMND paid their suppliers in advance. Future expenses is shifted to an earlier period which will inflate earnings.”

    The momentum payments are either:

    1. Payments for the 2010 crop but recorded in the 2011 crop year, allowing them to shift costs by 1 year and move those costs until after the Pringles deal is completed.

    – or –

    2. The momentum payment was an advance on the first payment to be given for the 2011 crop, which was harvested about 10 days later than normal.

    What makes people think it is #1 and not #2 are that Diamond was about 40% below the market price for its payments on the 2010 crop, which makes it look like the momentum payment was part of 2010’s crop payment. Also, growers who were not going to deliver to Diamond in 2011 and told them so were told to go ahead and cash their checks.

    Their contracts state that Diamond can set the price, but they also use the term “in good faith” and paying slightly more than half what the other 60+ packers in the industry pay does not sound like it is making payments in good faith.

    Disclosure: I am in the walnut business and privvy to the fine details of how the industry functions. I don’t have any stock or positions in Diamond and wouldn’t touch their stock with a 10 foot pole or even buy it with somebody else’s money.

  • Pacioli

    “I’d like to see an example of a company with huge accruals over several periods (80%+) to see how well it performed over time.”

    I went ahead and performed this exercise for my favorite stock, CF Industries (CF). I utilized the ‘Balance Sheet Accruals Ratio’ that you describe in this post: http://www.oldschoolvalue.com/blog/valuation-methods/you-need-to-determine-earnings-quality-through-accruals/ (despite some flaws in that evaluation metric that we could get into).

    What I found is that CF’s BS Accrual ratio has been all over the place through the years. For FY 2004, it was -30.8%. For the most recent year for which there is annual data (FY 2010), the ratio was 146.3% (well above your 80% threshold). Yet CF has returned 517% over the last 5 years, or 39% avg annual return. It’s been one of the best stocks I’ve ever come across.

    This empirical exercise merely confirms the suspicion I already had, regarding your contention about accruals. Namely, the mere existence of accruals (even if sizable) really has very little at all to do with analyzing a stock’s long term return potential. All accruals mean is that the company in question does not operate on a cash basis.

  • @ Pacioli,
    CF is a bad example.

    Unlike DMND, CF has increased cash, virtually zero short term debt and I dont know what the huge increase in long term debt and total liabilities for 2010 is.
    Also, unlike DMND, CF accruals have been perfectly fine and reasonable up to 2010. If their 2011 accruals are up in the 100% range for 2-3 years in a row, then that requires further digging. So what would my conclusion be for CF? Nothing. It’s fine.

    The accrual method isn’t meant for determining the long term return potential. It’s a systemized check to see whether future problems could arise.

    Find me an example where the actual balance sheet and cash flow accrual ratio using the formulas given is greater than 80% for 2-3 years and then we’ll continue this.

  • @ walnutexpert,

    Good information. Thanks. Only something an insider in the industry would know.
    Only increases my growing suspicion of accounting shenanigans.

  • Pacioli

    “Find me an example where the ratio is greater than 80% for 2-3 years and then we’ll continue this.”

    Alright, so even though you have not bothered to answer any of the questions posed in my original comment, I went ahead and performed the exercise again, as you requested in your recent comment quoted above. This time, I looked at a company I have been interested in for some time – SolarWinds (SWI). It is an enterprise software company.

    For the fiscal years of 2007-2010, its BS Accrual Ratio has been 989.6%, 87.5%, -112.1%, and 163.5%, respectively. So, 3 of the past 4 years exceed your 80% threshold; and all 4 years exceed it on an absolute value basis.

    Meanwhile, SWI boasts returns on capital in the low-to-mid twenties, and the stock has appreciated 68% over the last year.

    I think it’s safe to say we can agree to disagree on the usefulness (or blatant lack thereof) of this method as a “systemized check to see whether future problems could arise”. Based on the exercises I have performed here, at your request, the evidence is pretty clear to me.

  • Drew

    Not questioning there might be some shady stuff going on… but I’m not so sure about the concept of shifting earnings into an earlier period automatically means it’s wrong. After all, isn’t allowance/provision for loan losses and other similar write downs theoretically shifting future potential losses (expenses) into an earlier period?

  • @ Drew,
    Not exactly because the whole concept of allowance/provision of bad loans is to build up a small balance to offset any unforeseen writedowns that is not the fault of the company. If done incorrectly, a company could draw money and use it like a small bank account to inflate revenues, but this isn’t the case with DMND. The company purposely paid money in advance when it shouldnt have, thus increasing the expenses in that year, but making the following year much better than it should be.

  • @ Pacioli,
    A company with a very short history with huge accruals to start off. A bit too early to jump to conclusions don’t you think?
    Good thing you bring up SWI because there is a high chance that it will end up just like FSLR. If you want to determine long term value, as you mentioned, SWI is another bad example.

    So take a look at FLSR and see what happened. That should answer your question.

    Plus, if you don’t find this method useful, by all means, don’t use it. No one is forcing you.

  • walnutexpert

    @ Jae Jun,

    You state “The company purposely paid money in advance when it shouldnt have, thus increasing the expenses in that year, but making the following year much better than it should be.”

    What happened is they made a payment to the growers in September, which was their new accounting cycle, but growers content that it was a payment for the previous season’s walnuts. In essence, the claim is that they made payments for one crop year in the following accounting year to hide the true cost of goods and inflate their profit margin for another year, keeping a high stock price.

    The correction would have come in the future once the Pringles deal went through and may have even been hidden in the costs of merging the companies, allowing them to purchase a company with shares of Diamond that were artificially inflated due to accounting manipulation, effectively pump and dumping their way into ownership of a large portion of a much more valuable company.

  • walnutexpert

    Just one note on the intangibles – I thought that was crazy that 2/3 of their assets are intangibles and goodwill so I did some research.

    JBSS is another nut company and similar to Diamond in many ways. Their goodwill and intangibles are only 4% of their assets.

    Johnson & Johnson is 31%.
    Proctor & Gamble is 65%
    Kraft is 66%.
    Diamond Foods is 67%.

    Although it sounds high, other companies are up at those levels as well.

  • Pacioli

    “A bit too early to jump to conclusions don’t you think?”

    Not really – 4 consecutive years is enough for me.

    “Good thing you bring up SWI because there is a high chance that it will end up just like FSLR.”

    Why in the world would SWI have anything to do with FSLR? SWI is an enterprise software company. FSLR is a solar company.

    “…SWI is another bad example”

    Of course it is. Just like the first one I provided. If it dilutes your contention, then surely it is just “a bad example”.

    “…if you don’t find this method useful, by all means, don’t use it. No one is forcing you.”

    I have not and would not use this method, for reasons that are now obvious to the objective reader. My intention was to inform readers who may not know better. In my opinion, it is a disservice to recommend analysis techniques that are useless at best, and harmful at worst.

  • @ walnutexpert,
    Good catch. Very true.

  • Pingback: Weekly links « valueandopportunity()

  • DMND restates 2 years of financials, CEO and CFO placed on administrative leave to be replaced.

    I’m going to conclude that DMND is a good case study now that accounting fraud has been proven.

  • Pingback: Diamond Foods–Uncovering the Accounting Foolery | ShortZilla()

  • richard

    How can it be that accounting that decreases a former period’s profit and increases a later one’s profit is out of line? It all balances out, doesn’t it? Adding to one, taking away from another? How does that inflate anything beyond what it would have been if these periods were combined? Also, it seems to me that paying supplier expenses in advance is a great way to insure good future pricing and goodwill from those same paid suppliers. Wouldn’t everyone like to get paid in advance?

    Unless there is true accounting fraud where numbers were actually cooked/invented, I do not see the big deal here. I pay suppliers ahead of delivery all the time to get better pricing and to insure that they will always put me first in line. Other than the above, if someone can better explain the significance of what the Diamond C-suite did to harm Diamond, please do so, and thanks.

  • from just a private business point of view, it isn’t as bad, but for a public company it is direct manipulation of results. By shifting expenses from one period to another, they are deliberately trying to make a certain period look better than it really did. E.g. there is a big difference in earnings $2 EPS in 2010 and $3 EPS in 2011 compared to $1 EPS in 2010 and $4 EPS in 2011.

    What do you mean C-suite?

  • Pacioli

    Interesting choice to selectively block certain commenters’ access to your site. Seems very “Chinese gov’t” to me, but to each his own, I suppose.

  • Carl

    “In my opinion, it is a disservice to recommend analysis techniques that are useless at best, and harmful at worst.”

    Pacioli,
    I read the original blog again and where did Jae say I recommend this analysiss technique to the readers ?

  • @ Carl,
    Thanks but it’s ok. No need to respond to non-constructive comments.

  • Testing

    @ Carl – I guess I would point to “looking at accruals along with other manipulation detect [sic] techniques will help you identify fraud”.

  • walnutexpert

    @Richard “How can it be that accounting that decreases a former period’s profit and increases a later one’s profit is out of line? It all balances out, doesn’t it?”

    Well, yeah, it balances out, but if the later period hasn’t occured yet and the accounting maneuver is hidden, then it is fraudulent, no?

    I made $3 million this year. I made $2 million last year. I am rich. I do not tell people I will lose $5 million this year because I actually didn’t make anything but I shifted expenses into the future to make myself look rich. Now, I want to take out a loan on my income of $2.5 million per year and use my historical income as evidence of my credit worthiness.

    Would the above not be fraudulent because it will all work itself out eventually? Certainly not!

  • pravchaw

    DMND was trying to use its inflated stock price as currency to acquire Pringles from P&G shareholders. Fortunately for the latter this fraud did not progress.

  • There is growing suspicion that insider trading occurred before the news was announced. Getting more interesting each day.

  • Pingback: The Complete & Practical Guide to Mastering the DEF 14A Proxy()

  • Pingback: Stunning Growth but Shady Business()

  • Joy

    I am just confused. If DMMD treated the momentum payment as an advance, isn’t that a liability? How can that be recognition of the cost in the later period? Aren’t those costs the costs for future cost originally?

[checklist-collection]
[checklist-collection]