(To read Part 2 of this series, click here)
In light of the recent attacks on NQ Mobile ("NQ", or the "Company"), we intend to publish an indefinitely long series of refutations to provide investors with a balanced view.
It would take us at least the 81 pages that Muddy Waters ("MW") wrote to do a point by point refutation, so we will initially focus on core elements of the short thesis, which are summarized below:
- NQ had no cash at the time of its convertible note issue because it reclassified all its cash and equivalents to level 2 in their 2012 20-F filing
- NQ's payment service provider, Tianjing Yidatong ("YDT"), is an off-balance sheet shell company used to roundtrip cash and manufacture revenue because YDT's SAIC financials don't tie exactly with NQ's financials when a litany of curious assumptions are taken as true
- NQ can't possibly have the domestic market share it claims because some surveys MW did lead to a different conclusion
- NQ's security software is actually spyware that insecurely sends personal information to the Chinese government. NQ's enterprise, gaming, and advertising businesses, not to speak of its recent innovations such as the popular Music Radar search app and the recently unveiled NQ Live platform, are frauds too (mainly by association, we suppose)
Ergo, with no cash, no real revenue, no customers, a spyware product, and a bunch of other fraudulent businesses, NQ must then be a zero. Strong Sell!!
We just spared you 81 pages of reading. That's basically the hit job in a nut shell.
An investment thesis is only as good as the premises upon which it is built. The strongest theses are supported by facts, while weaker theses are supported by assumptions. Then, there are fraudulent theses that are built upon fallacies, such as the MW short thesis on NQ.
In the interest of time, we will address the core elements to the short thesis in parts. In this Part 1, we analyze the most severe allegations, #1 above, which question NQ's cash. We will publish our thoughts on #2 tomorrow.
Before we share our views, for the benefit of the uninitiated, a little Chinese ADR history is in order. After all, when fraud is in question, the credibility of the short seller is as important a consideration as the credibility of the company in question.
I. A Brief History of Chinese ADR Short Selling
The prevailing sentiment in US markets is that Chinese ADR issuers are not to be trusted, that frauds are so prevalent that blowup risk far outweighs reward that the short sellers who were seemingly often right in the past will continue to be right.
Since the crowd is usually wrong in the long run, let us consider the contrarian perspective. Are the shorts who were right in the past as likely to be right now? Or are they the Jeremy Lins of the investing world, hot one year, only to fade the next when the Linsanity ends?
The Golden Years
It is hard to erase the recent memory of 2011, when over 20 Chinese frauds were uncovered. Literally dozens of short reports a month flooded the market proclaiming fraud. As Jim Chanos added fuel to the fire by calling the entire country a big pile of steaming fraud, 2011 was the year of manifest destiny for short sellers. The mere mention of the possibility of a bear raid sent stocks reeling -10% any given day.
A fairly exhaustive analysis of the history of Chinese frauds can be found in an academic paper authored by Charles Lee, an accounting professor at the Stanford Graduate School of Business, et al. titled "Shell Games: Are Chinese Reverse Mergers Inherently Toxic?" Lee shares a much shorter summary of his conclusion (that Chinese reverse mergers are actually good investments) in a Bloomberg interview held January of this year.
(For the record, we point out Lee's work primarily as a reference for data and do not necessarily agree or disagree with any of his conclusions.)
As Lee notes, the Linsanity has settled down considerably since 2011. Why?
The vast majority of Chinese frauds had several things in common. They tended to be micro-caps, they were rarely audited by reputable accounting firms, their sell-side coverage and corporate access were effectively non-existent, and, of most importance, they overwhelmingly employed reverse mergers (RMs, AKA reverse takeovers, RTOs) to gain access to the equity markets.
RMs, when used properly and ethically, can be fabulous vehicles for the creation of shareholder value. After all, the world's most successful reverse merger ever is a household name: Berkshire Hathaway (NYSE:BRK.A) (NYSE:BRK.B).
But when used unethically, RMs are the ultimate fraud-ing machines. First, they do not require the intense scrutiny of an IPO process. A private company buys enough shares of a publicly listed shell company to control it, then exchanges its shares with the shell's shares, and voila, it is officially public. No IPO audit, no road show, no SEC-reviewed prospectus replete with disclosures, basically, almost nothing. For fraud purposes, as important as bypassing the IPO is having relatively low liquidity. The less people you have to dupe, the higher your chances of duping. And the lower the liquidity, the easier it is to manipulate prices.
Is it any wonder, then, that the vast majority of Chinese frauds were public via RMs? Use of this loophole reached a tipping point a few years ago until the fraud bubble finally burst in 2011.
Muddy Waters is the most famous, if not exactly the most successful, China-focused fundamental short seller, so its track record is particularly interesting. (We exclude the notable Jim Chanos of Kynikos Capital who, in our opinion, invests fundamentally but with an overarching macro thesis.)
Below is a summarized history of MW attacks.
Spotting the salient trend does not require a PhD in statistics.
MW has not gotten a single non-RM call right yet. To us, the obvious trade appears to be getting long after a MW hit job. (For those willing to cross the ethical line, becoming a MW client and front-running the hit jobs for an intra-day trade appears to work as well.)
Without the RM loophole, it is much, much harder to defraud investors. Statistically, the likelihood of fraud drops precipitously when an IPO was involved. And it drops precipitously again from that point when market cap (a reasonable proxy for liquidity) exceeds $1 billion, as was the case for NQ prior to last Thursday.
Market sentiment may still favor the shorts, but with rampant RM fraud largely out of the picture, probability of fraud overwhelmingly does not.
II. Level 2: The Massive Red Herring Behind the Alleged "Massive Fraud" at NQ
The most important argument to MW's short thesis is that NQ's balance sheet has no cash other than the cash recently raised via the convertible note issue. MW claims that Big 4 auditor PWC was negligent (quite the claim, since there are severe legal consequences for PWC if true), that they did not check the cash deposits that were allegedly diverted by the Company and falsified on the balance sheet.
This is certainly a scary pronouncement, since it calls into question the most fundamental measures of value: cash and cash generation.
There are two separate allegations to explore here: that NQ has no cash, and that PWC's sloppy audit provides investors no comfort.
MW arrived at their conclusion that NQ must have no cash by reading the Company's 2011 and 2012 annual 20-F filings and noting that cash and cash equivalents were reclassified in 2012. Knowing that 99.9% of investors will not be intimately familiar with audit accounting, MW employs a confident tone and references to accounting esoterica (the classification of cash assets) to mislead investors into believing there is a problem in NQ's balance sheet.
Specifically, MW claims that the reclassification of NQ's cash assets from level 1 to level 2 is problematic. Their theory is that actual cash, as opposed to cash equivalents such as term or time deposits and CDs, can only be classified Level 1, so the absence of any Level 1 assets on the balance sheet means that there is no cash. Since any business requires cash to fund operations (day to day expenses, working capital, etc.), and all of NQ's cash assets are level 2, NQ cannot be a real business and must therefore be a fraud.
Based on the stock price reaction to these claims, it is not hard to guess what the market thought, but to those familiar with audit accounting, MW's analysis is laughably naïve and incorrect.
The major problem with MW's line of thinking is that it assumes the following fallacies:
1. Cash must always be classified level 1, no exceptions
2. Level 2 assets are not liquid enough to be used as actual cash
3. The reclassification of cash assets points to a major problem
Fair Value Measurement
Any full audit of a company's balance sheet will include an assessment of the value of its assets. Because asset valuation is part art, part science, accounting firms usually make disclosures as to how they are valuing the assets. For cash and equivalents, this disclosure is sometimes as simple as a one-liner that states something to the effect that "all cash equivalents are fairly valued." While that is not much of a disclosure, not much of one is usually needed, since, cash being king, investors tend to be satisfied that if an auditor checked the cash, the cash is "marked" properly.
Some accounting firms opt for more detail, categorizing cash and equivalents into levels 1-3.
The most important point to remember about these levels is that they are nothing more than disclosures of how value is measured. They are NOT measures of asset "quality," they are NOT measures of liquidity, they are NOT measures of volatility, and they are NOT measures of risk. They are just disclosures of valuation methodology.
Below is a standard explanation of what these levels mean in the "accounting-ese" used in a 20-F filing:
In plain English:
- Level 1: securities that can be actively traded and is therefore valued directly (a share of NQ or FB or MSFT is a level 1 asset)
- Level 2: securities that can be valued indirectly, e.g. by comparing value to a similar asset that is itself directly valued/traded (a CD with a fixed maturity is no longer actively quoted once it is bought, but it is easily valued because there are other CDs with the same maturity to compare it to)
- Level 3: applies to hard to value assets (such as private equity funds, venture capital funds, etc.)
One of the positives from all the Level 2 hullabaloo is that he impeccably credentialed Dr. Paul Gillis, who literally wrote the book on the accounting field in China, has jumped into the fray.
Here is his opinion.
So it turns out, the foundation of MW's allegation that NQ has no cash is based on a rookie accounting mistake.
Interestingly, as Gillis alludes to in his blog, cash money itself is not clearly a level 1 asset! Some have argued that it is in fact a level 2 asset, since it is not actively quoted/traded. We immediately assume it is Level 1 because one could go to any bank and get $1 for $1. But it is not actively quoted in a market like, say, a stock.
The reality of fair value measurement is that the classification of assets into Levels 1-3 is a relatively subjective exercise that varies widely across audited companies and auditing firms, and even with the various teams within the same auditing firm.
- Not only do we agree with Dr. Gillis when he opines that NQ is actually leading their ADR peers in the correct implementation of latest FASB standards, we note the following facts about NQ's cash balances that we recently learned through interviews with several partners of Big 4 accounting firms and Chinese CFOs of our portfolio companies:
- In China, especially at the larger (state-owned) national banks, large depositors such as NQ often get the benefit of being able to liquidate term deposits with relatively lengthy maturities (such as NQ's multi-month or even one-year term deposits) within 24 hours. Of course, a penalty commensurate with the interest accrued from the higher-yielding security must be paid, but these assets can be practically as good as cash from a liquidity standpoint. If it wanted to, NQ could liquidate all its term deposits and convert its entire net cash balance to cold hard cash money tomorrow (but why on Earth is that desirable?).
- NQ disclosed all its term deposit assets as proof of value and liquidity to investors on Friday. However, contrary to MW's opinion, the Company does also have actual cash deposits in local banks near some regional offices, such as the Dallas co-HQ, for ordinary day-to-day uses such as making payroll, paying rent, buying pizza for their hard-working engineers, etc. Those cash balances, which total roughly $20-30 million, were not disclosed, probably because the sums in each account were significantly smaller than the size of the accounts holding term deposits.
- Some firms value cash and cash equivalents based on value-based formulas that assign classification to just one level if the significant majority of cash assets is of one level or another. This could explain why certain accounting firms, even if operating under the premise that cash by itself must always be Level 1, might lump it into Level 2 along with cash assets such as term or time deposits.
- NQ's 2011 20-F includes an important footnote below the fair value measurement table (see image below) that discloses a year ahead of the FASB changes that NQ will reclassify its cash assets in 2012 to conform to the latest disclosure standards. So the notion that the reclassification came out of nowhere and is suspicious could not be farther from the truth.
Peer Company Accounting Practices
We screened the 2012 annual reports for all constituents in Bloomberg's Chinese ADR index (all 55 of them). Our key findings follow:
- First, the whopper: Despite all the Level 2 noise, 69% of the index constituents did not even report categorization for their cash assets! The vast majority did not specify Level 1-3! How significant does this Level 2 "problem" seem now? A few examples include some of China's largest and most respected names: CTRP, CHL, et al.
- KPMG did not categorize the cash equivalents of any client
- Deloitte categorized the cash equivalents of only three clients
- E&Y categorized the cash equivalents of only half of its clients
- Contrary to MW's suggestion that NQ's level 2 cash is unique, there are, in fact, other examples of companies accounting the same way, such as CTRP competitor LONG
- Reclassification of Level 1 assets to Level 2 is not as uncommon as MW would like you to believe. BIDU, among the most respected Chinese companies, reclassified RMB 2.4 billion of time deposits from Level 1 to Level 2, for example
Again, as the new FASB rules gain tenure and the various accounting firms learn to comply with them properly, there will be wide variance in the categorization of cash assets, which could take years to sort out into what may be deemed as an industry best practice. Today, that does not exist.
In conclusion, the key assumption that MW uses to "prove" that NQ has no cash is based on an accounting technicality that MW does not even properly understand in the first place. The entire foundation of the argument is false.
Now, on to the likelihood of accounting fraud.
Recall that the second allegation MW makes about NQ's audited financials is that PWC was asleep at the wheel. Given the insanely high scrutiny of Big 4 auditors over the past two years, especially this year as the US and Chinese governments negotiate a way to provide better disclosure of Chinese working papers to US investors, we find this allegation to be highly improbable.
But even if PWC were asleep at the wheel, NQ might just be the most audited company in its peer group. Despite being a relatively young public company, having been listed for just 2.5 years, below is a list of audits and due diligence investigations that NQ has passed with flying colors (each of these included cash verification, business contract verification, and channel checks):
- 2011 IPO audit (PWC and SEC)
- 2011 20-F filing (PWC)
- 2012 20-F filing (PWC, though asleep at the wheel this year, per MW)
- 2013 Atlantis PIPE transaction (unnamed accounting firm, Macquarie)
- 2013 Macquarie due diligence (initiation of coverage)
- 2013 Wedge Partners independent local filing document review (same SAIC filings used by MW, though cash was not verified this time)
- 2013 Morgan Stanley due diligence for convertible note offering
- 2013 Deutsche due diligence for convertible note offering
- 2013 Convertible note offering (due diligence by all participants of offering)
- 2013 Post-Attack cash verification by two unnamed investment banks (as announced by Khan on Bloomberg TV last week)
- 2013 Post-Attack cash in-person verification by multiple large funds (as announced by Wedge Partners)
If NQ has no cash, then that is certainly a lot of people who were asleep at the wheel.
We know for sure that at least two of them were not asleep at the wheel, since the two unnamed investment banks (we guess, Morgan Stanley and Deutsche Bank, at the request of those who participated in the convert offering…) returned to NQ's Beijing headquarters after their bank field trips with the following digital camera snapshots of deposit slips for every term deposit NQ has ever made to the various bank accounts disclosed to investors the day after the MW attack last week.
As promised during the investor call last week, when we asked for the snapshots that validated the bank account information from the post-MW press release, NQ delivered. We post them here, here, and here for the benefit of fellow investors.
If that is not enough to convince that NQ is not actually a fraud, perhaps the $100 million of cash that the Company announced today will be wired to Standard Chartered, a global leader in commercial banking, will be reassuring.
Interestingly, the shorts seem to be retreating given this news. The Netease article referred to in this post that tanked the stock yesterday has now been removed. One wonders why.
Additional disclosure: Refer here.