Words and numbers
One of the best chapters in Steven Levitt and Stephen Dubner’s Freakonomics is a close analysis of real-estate terms – such as charming (which means the house might be an acquired taste), great neighbourhood (the house itself is terrible), well-maintained (old but not yet falling down) and fantastic (definitely not fantastic).
Everyone knows that being a real-estate agent is about persuasion, but Levitt and Dubner argue estate agents are as concerned with persuading the seller to sell as they are with convincing the buyer to buy. A sale is more likely, and the seller’s commission only marginally reduced, if the seller adopts a lower reservation price.
Levitt and Dubner depict agents using particular words to signal useful information about properties and their sellers’ expectations. A critical finding is that some words – such as granite, state-of-the-art, maple and gourmet – are strongly and positively correlated with the ultimate sale price.
Which brings us to Blue Orca Capital Texas, LLC, Glaucus Research Group California, LLC, and the strange world of activist short-selling.
The medium is the message
In the global financial eco-system, the activist short-seller is an exotic species: a formidable hybrid of a conventional short-seller and an ordinary think-tank or research outfit. To understand what an activist short-seller does, let's take a quick look at conventional short selling.
Conventional short selling is a kind of bet: a gamble that a share or some other type of asset will fall in price. The aim is to 'sell high' now an asset that you don't own, with the idea of 'buying low' the same asset later. How is that possible? It only works if you can 'borrow' the asset and then repay the 'loan' by delivering the asset at a lower cost in the future. Not all securities are approved for short selling. In Australia, the shares you can 'short' are typically the frequently traded ones and those of larger companies.
When 'shorting' is allowed, the potential payoff for the short-seller is the difference between the current and future values of the asset. Short-selling can make sense if you have good grounds to think a share is overpriced. In principle, it is also a way to insure yourself against a general fall in share prices.
But the danger with short selling is that there is no limit on the cost of repaying the 'loan'. If, instead of going down, the share price goes up a lot, the short seller faces much pain in closing out the short position.
Activist short-sellers take all of this one step further. They 'short' shares that are vulnerable to negative research. And then they publish research that has what they hope is the right level of negativity.
There are precedents for this business model, including the short-seller Muddy Waters, but a new breed of brash and smart short-sellers - firms such as Blue Orca Capital and Glaucus Research - have taken the model to a whole new level.
Glaucus Research began operating in 2011. Its co-founder and research director, Soren Aandahl, studied law at Harvard and seems to have been instrumental in developing the new activist approach. In 2018, Aandahl left Glaucus to set up Blue Orca Capital, which operates very much in the Glaucus mould.
Several of the innovations adopted by Blue Orca and Glaucus are verbal. Their reports use especially sharp and racy language, such as 'nosebleed prices', 'gimmicky fad' and 'wild outlier'. Crucially, their reports are available fee-free on the internet.
In 2017, Glaucus Research began targeting companies in Australia. (Corporate law firm Gilbert + Tobin described Glaucus's campaigns as ‘unorthodox and aggressive’.)
Glaucus’s first foray in Australia focused on TFS Corporation Ltd (later ‘Quintis’), a WA-based grower of Indian sandalwood. In March 2017, Glaucus published a report that said TFS’s liabilities substantially exceeded its assets; its corporate disclosures were misleading; it would struggle to access capital markets; and it was therefore headed for bankruptcy. (I'll return in a moment to whether these statements were true.)
Glaucus’s second Australian target was the Queensland-based asset manager, Blue Sky Alternative Investments Ltd. In March 2018, Glaucus claimed Blue Sky had overstated its fee-earning assets under management, and that Blue Sky shares were more than 75 per cent over-valued.
Much research in accounting and finance is about the terminology used to report business performance. As Loughran and McDonald observed in 2016, there is an inherent challenge in studying financial disclosures: ‘the problem of separating the business and the document. These issues are intertwined because the document attempts to describe the economic reality of the business’. With activist shorts, though, the situation is somewhat different. In an important sense, the activist’s business is the document.
Wild, ridiculous, ludicrous
Glaucus’s two Australian reports together serve as a useful case-study of the impact of words in finance. The reports tell vivid and compelling stories of ‘massive conflicts’, ‘festering uncertainties’, ‘wild exaggeration’, ‘ridiculous projections’, ‘highly misleading statements’ and ‘absurdly optimistic assumptions’. A close reading of the reports reveals much of linguistic interest.
First of all, the venerable art of rhetoric is key to the activist toolkit. Glaucus strikes several rhetorical poses and deploys multiple sub-rhetorics. The most prominent of these is a rhetoric of science and investigation. Keywords repeated in the Quintis and Blue Sky reports include discovery, uncovering, close look, sceptical and evidence.
This rhetoric is reinforced by the report framing – the reassuring disclaimer up top, share prices and other data down the side like an old-fashioned broker’s report – and by the emphasis on quantification and precision. Rather than say ‘the Quintis shares are worthless’, for example, Glaucus says: ‘we value [its] equity at AU$0.00.’ That’s zero to two decimal places. Bracketed asides build intimacy and trust between author and reader: ‘(and for the record, we doubt it)’, ‘(supposedly)’, ‘(pun intended)’.
In the two Australian reports there’s also a rhetoric of crime. The following words appear in both reports: suspicion, unjust, gouge, abusive, egregious and distort (the Quintis report also includes the dreaded P-word: Ponzi). There’s a rhetoric of emergency and imminent danger, evident in common words such as collapse, burn, red flag and balloon (about to burst). And a rhetoric of artifice and dissimulation: mystery, dubious, opaque, absurd, unusual, ludicrous. There’s also a ‘dad jokes’ rhetoric of hokey and hackneyed phrases: the shoe fits, do your homework, the tip of the iceberg, winter is coming.
Taken together, these rhetorics coalesce into a compelling gestalt of honesty, science and revelation.
Thrusters, scoopers and scroungers
The conduct and performance of large businesses is important. It affects our wealth and health, the health of our environment, and much else besides. But there’s a problem here. Customers, concerned citizens, small shareholders and potential investors all demand knowledge about what companies do, but finding things out is often difficult. Corporate structures at a fundamental level are about confidentiality. Their legal DNA establishes an information barrier.
Different generations have pierced this barrier in different ways. Mandatory disclosures. Mandatory audits. Regulatory oversight. Protections for whistle-blowers and investigative journalists. Corporate Facebook pages and Twitter accounts. All these are different ways to find things out and hold companies to account.
In a 2010 journal article, I attempted to classify the different kinds of disclosure and accountability mechanisms. Companies making proactive disclosures were ‘thrusters’ because they sent information outward. Auditors were ‘scoopers’ because they entered organisations, gathered up information and then took it away. Bounty hunters and citizen auditors were ‘scroungers’: they circled the corporate perimeter and did what they could with information leaks, ‘open book’ disclosures and other scraps of data that could be obtained from outside.
The reports of activist short-sellers are mainly not about disclosure of private information. For the most part, their inputs are already available in the public domain. The short activist’s business model is about gathering and repackaging information to tell a story. In my taxonomy of accountability mechanisms, the activist short-seller is a special kind of scrounger.
Why special? Because, by deploying investigative rhetoric, activist shorts adopt a posture that is more scooper-like than scroungy. And because, though they are a type of bounty hunter – a classic variety of scrounger – both the bounty and the hunt are unusual. For activist short-sellers, the stockmarket rather than the law provides the bounty. And the bounty’s size is ‘endogenous’, determined by how well the activists’ reports change people’s minds.
In the accountability taxonomy, activist short-sellers have a lot in common with investigative journalism. They use many of the tools in the journalist’s toolkit: they cold-call front-line staff; they scour social media; and they do basic research to verify what company executives say about their assets, their profitability, their experience and their qualifications.
But there's an important difference. Activist short-sellers use these journalistic tools without the journalistic ethos of impartiality and disinterestedness. There are good reasons for this. The activists’ research reports might look like the work of investigators and advisers, but they are far from neutral. They aren’t instruments of disinterested analysis. The whole point of the activist short-selling business model is that the authors of the research are utterly partial and very interested indeed.
Like real-estate advertisements, short-sell reports are tools of persuasion. They are pieces of marketing, or anti-marketing. By publishing their reports, short-sellers are trying to change people’s perceptions of value – because perceived value is the basis for share prices, and price levels determine the short-sellers’ profits.
(The truthy rhetoric of the Glaucus Research reports is somewhat at odds with their fine print which states, among other things, ‘You are reading a short-biased opinion piece. Obviously, we will make money if the price of [the] stock declines. This report and all statements contained herein are the opinion of Glaucus Research Group California, LLC, and are not statements of fact’.)
The activist short’s business model is as fascinating as it is controversial. It is new in Australia, and our financial system is yet to catch up. Companies on the receiving end of short-sell word-bombs have called them vicious and irresponsible. Short-sellers in their turn speak of receiving death threats. But the proof, a Glaucus report might say, is in the pudding. How effective were the reports? What happened to their targets?
Outside Australia, some activist reports have misfired; the impact on the share price was too small, or prices moved in altogether the wrong direction. Inside Australia, Glaucus imitators have entered the field – and they’ve had a mixed record of success. But the two Australian Glaucus reports were devastatingly effective.
After the second report was published, Blue Sky’s market value fell by around $800 million. International investor Oaktree subsequently rescued Blue Sky with a $50 million loan, on terms reportedly favourable to Oaktree. Glaucus, it seems, was broadly right about Blue Sky. And it was even more right about Quintis: the sandalwood grower’s shares were indeed worthless. In January 2018, Quintis went into receivership; its shareholders are likely to get nothing, to two decimal places.
Company executives around the world are right to dread visits from activist short-sellers. Blue Orca, Glaucus and their peers emphasise they’re doing nothing wrong and that they will readily defend their reports in court. The activists lob intelligently crafted pieces of financial propaganda, but they are also part of our corporate accountability architecture, and they seem to be evolving in interesting directions.
Glaucus’s two Australian research reports have a lot in common with each other but they are far from identical. The second report is much longer. It introduces new keywords and phrases, and uses some old ones with greater frequency. It applies the same varieties of rhetoric but with different emphasis.
Is this evidence of short-sellers becoming more sophisticated? In deciding what words to use, are the activists choosing the ones with the greatest likely impact? Are gouge, ludicrous and Ponzi the short-sell counterparts to the estate agent’s granite, maple and gourmet? Are the activists optimising their use of these words, or could astute linguistic changes further increase the reports’ power? And could digital tools such as cognitive AI do a better job of word selection?
There are also important unknowns for investors and policymakers. How might investors tell which short-sell words are most informative about the condition of the targeted companies? What implications do the short-sellers’ tactics have for our systems of corporate accountability and disclosure?
The intersection between economics and linguistics can help answer these questions. It’s a small field with a big future.