The Book of Family Traditions on The Art of War, Yagyu Munenori
Earlier yesterday I was on BNN to talk about Valeant's alleged accounting manipulation (http://www.bnn.ca/Video/player.aspx?vid=736200). Valeant is the result of a merger between Biovail and Valeant, both of them had an accounting scandal in their history. I actually use Biovail as a case in my courses as an example of revenue manipulation, so one could argue that there is genetic tendency there.
The story here is whether Valeant has created specialized drug companies (remember Special Purpose Entities or SPE's in Enron) to book fictitious revenues. The company claims that it consolidated them, which could refute this claim. If this is true my question is why would employees of Valeant use pseudonymous like Peter Parker (seriously?) or Jack Reacher in emails from Philidor, one of such companies created by Valeant?
When one goes beyond simplistic heuristics and uses the Beneish Manipulation Index analysis (below) we can see that while the overall probability is between 1% and 2% (not so low as you'd think) the first index is highly suspicious:
Weighted Predictor Ratios Annual | 2012 | 2013 | 2014 |
Days Receivables Index | 1.01271 | 0.984018 | 0.792907329 |
Gross Margin Index | 0.507612 | 0.56891 | 0.487203384 |
Asset Quality Index | 0.382818 | 0.402746 | 0.391145484 |
Sales Growth Index | 1.300824 | 1.529888 | 1.281557525 |
Depreciation Index | 0.100797 | 0.126199 | 0.125124344 |
Sell. & Admin. Exp. Index | -0.155772 | -0.173113 | -0.185863 |
Leverage Index | -0.395363 | -0.334107 | -0.317103 |
Total Accruals/Total Assets | -0.20139 | -0.31876 | -0.245703794 |
Constant | -4.84 | -4.84 | -4.84 |
Value of y | -2.28776 | -2.05422 | -2.510731907 |
Probability of Manipulation | 1.11% | 2.00% | 0.60% |
The Days receivable Index is related to whether accounts receivables grow faster than sales. If they do, unless the credit policy has changed, it could potentially be the result of channel stuffing or booking of low quality sales. What is even more suspicious is the exponential growth of sales and that the rate that receivables grow is even greater than the rate that sales grow. So what we get here is that sales grow very rapidly but receivables even more.
Thus, when I looked at the growth in sales vs. receivables I got:
2012 | 2013 | 2014 | |
Sales growth | 46% | 72% | 44% |
AR Growth | 61% | 83% | 24% |
When I charted Valeant's quarterly accounts receivables and sales I got the following stunning result:
The sales in each quarter from 2011 and the receivables are almost the same. As a matter of fact the correlation between the two series is 98%.
What can we conclude?
The sales in each quarter from 2011 and the receivables are almost the same. As a matter of fact the correlation between the two series is 98%.
What can we conclude?
That the overall probability of earnings management is inconclusive but there are red flags with respect to the revenue recognized by Valeant.
Professor Elitzu, could the high correlation between sales and AR is just a matter of fact that Valeant sales is mostly based on credit and collect them next quarter? So very high turnaround of AR like collect AR and sale, and collect AR next quarter and sale...?
ReplyDeleteThis kind of details are lost in cashflow statement.
Could be. At the same time it does not explain why are they growing more than sales.
ReplyDeleteAlso, the question implies that the sales occur on one day during the quarter (presumably the last day) and then collected the next period. If this is the case this is even more damning. Usually sales and collections occur continuously throughout the quarter.
ReplyDeleteProfessor Elitzu, it appears that you are adjusting revenue and A/R. I assume for some of the recent acquisitions (Salix, etc.)? If so, can you provide some details around your adjustments? Thanks!
ReplyDeleteThis comment has been removed by the author.
ReplyDeleteProfessor Elitzu, in the period from 2012 to 2014 (inclusive), cumulative sales and A/R growth are 261% and 265% respectively. The corresponding values for the period Q2'14 to Q3'15 are 49% and 56% respectively. These numbers are not too far off.
ReplyDeleteCould the variations be due to the highly acquisitive nature of the company and the variation of the credit policies of the various acquired companies?
Unlikely, acquisitions require that the comparative numbers should be restated.
Delete