Herbalife (HLF): Is it a Fraud? Not likely.
Herbalife is currently a highly debated stock. On one side you have Bill Ackman claiming the firm is a ponzi scheme. On the other side you have Dan Loeb buying the stock with open arms.
One set of tools we describe in our book Quantitative Value, is how to apply statistical tools to identify manipulators, frauds, and/or potential bankruptcies.
“More money…has been stolen with the point of a pen than at the point of a gun.”
— Warren Buffett, Chairman’s Letter, 2000.
- Financial Statement Manipulation – financial statements fail to tell the whole truth about a company’s financial health/condition.
- Fraud – misrepresentation made that may result in unauthorized benefits to an individual, the firm, or a third party. Affected by opportunity and pressure.
- Financial Distress or Bankruptcy – when a firm has difficulty or cannot meet its obligations to creditors.
- Accrual measures (2 versions)
- Accrual Anomaly: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2598
- Net Operating Asset Anomaly: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=404120
- Manipulation prediction model: http://myweb.ncku.edu.tw/~r16001205/w7.1_ProbM_Model.FAJ.1999.pdf
- We apply 1 statistical tool to identify when a firm may enter financial distress/bankruptcy: http://w4.stern.nyu.edu/salomon/docs/Credit2006/J_Hilscher.pdf
What do the quant models say?
As of December 31, 2012, the quant model recommended purchasing Herbalife. The firm is very high quality and became excessively cheap after Ackman came out with his “short news.” My guess is Loeb bought our book over the holidays, read it, and then was determined to buy Herbalife.
How did the Fraud/Manipulation/Bankrupty models stack up?
- Accrual measures relative to universe of stocks
- Accrual Anomaly: 81 percentile
- Net Operating Asset Anomaly: 18 percentile
- Average: 49.5% percentile–basically, no issues
- Manipulation prediction model: Less than a 1% probability of manipulation; no red flags on any single manipulation metric
- Bankruptcy prediction model: The absolute probability of HLF going bust is low, but HLF scores at around the 89% percentile on this metric relative to the universe analyzed (stocks over $1.4B). This is something to watch, but the absolute probability of this occuring is very low (<1%)
Overall, the statistical results indicate that Loeb’s position is a better bet than Ackman’s position. Of course, this is in reference to the 12/31/2012 HLF stock price. As of yesterday, HLF is no longer included in the quantitative value screen because it has become too expensive.