Notes on “Selling Short”, Kathryn Staley

A video of the 2007 presentation is here. Disclaimer in case of potential takedown: this blog’s for fun, not profit, and I took notes out of respect & a desire to share.

—

Most true short-sellers are very moralistic.

• The best job a short-seller could have would be working for the SEC (except for the bureaucracy).

Paranoid, smart, individualistic, independent, and compulsive about their work.

Staley earned her CFA after business school and became a growth-stock investor.

• When her first company collapsed after mgmt lied to her about the business’s prospects.
  • She asked a CPA former classmate to help her go through the financial statements. He showed her that she could have forecast the collapse if she had tracked receivables.
    • She became a fan of accounting.

Kathryn honed her stock-picking skills under quality of earnings-focused analyst Ted O’Glove.

• Abraham Briloff was the only other writer focusing on earnings quality at the time.

The short-selling community was insular and wouldn’t speak to her, nor was there much written about the subject, so she wrote her book based on her case-by-case experiences shorting.

She describes her method of short-selling as accounting- & information-based, vs. arbitrage or quant.

The group of short-sellers espousing her philosophy consists of “maybe 20 people”, no larger today than it was back in the ‘80s.

She retired not long after writing her second book in the late ‘90s, although this talk invitation in ’07 got her interested and she conducted interviews to see how companies had evolved since Enron et al.

“A great short-sale candidate is a company with no cash flow, no assets, a bad business plan, huge float, high stock price, and very sleazy mgmt… but, of course, you can’t always get what you want”

• “Short-sale candidates are not negative earnings surprise companies… We’re primarily looking for companies that are going to go away.”
• “Most of the short positions I held lasted over 2 years… some lasted 10 years.”
• You have to be confident that the company is deeply flawed to hold a position that long.

Components of short analysis:

1. Quality of assets:

• • Assets that are “unbankable” will blow up and kill the company/earnings.
  • These are “fuzzy” assets or ones that depend on the welfare of another company.
• Example: Jiffy Lube (ticker: JLUB, now a subsidiary of RDS):
  • JLUB is an oil-change franchise.
  • Almost all assets depended on franchisees—A/R, LT & ST receivables, and centers held for resale.
  • As JLUB’s business deteriorated, number of poor-quality assets grew, until the company reached a BV of $80 M—with $180 M in assets dependent on franchisees.
    • Therefore, as a short-seller, you needed to watch the franchises.
  • JLUB took out an ad in the WSJ that read: “Entrepreneurs of all sizes, grab the Jiffy Lube opportunity. No money down for owner-operators.”
    • “What that ad meant was that Jiffy Lube had too much money.”
    • “If you owned the stock at that point, you certainly would have sold it if you were paying attention… and if you were short it, you would have added.”
  • Several years after she exited her JLUB position, Staley found out that a classmate of hers was Jiffy Lube’s banker. He admitted that his bank (and JLUB’s investment banker) was happy to loan as much as possible to JLUB.
    • Meanwhile, Jim Henman, Jiffy Lube CEO, was happy to lend out all of this money to potential franchisees.
    • Store growth was very high, while store quality was low.
• Example: Boston Chicken (ticker: BOST):
  • This was another franchise-fueled hypergrowth strategy gone wrong, with similar asset-quality issues.
  • However, the main clue was not on the B/S: an off-balance-sheet lease facility to finance franchisees.
  • The tell was that BOST’s return on invested assets was single-digit.
• Other unbankable assets are goodwill, deferred anything, securities not market to market, capitalized anything…
  • Grocery store chain J. Bildner & Sons once capitalized its Christmas party.
• Other B/S areas of interest include receivables & inventory.
  • Now, DSO & DIO are tracked closely by funds, so it’s not as useful anymore.
• Example: LA Gear (ticker: LAGR) was a disciplined shoe company, and inventory growth never exceeded COGS.
  • “The quarter when inventory was up a little bit more was the time to short the stock.”
    • Soon after, inventories grew +156% in a quarter when COGS rose +33%. The stock plummeted.
• Example: Career Education (ticker: CECO), a for-profit education company, was an interesting contemporaneous case she mentioned.
  • It was a hot growth stock with good cash flow and DSOs of only 20 to 30.
  • The company sold Title IV programs, or vocational training for high-school graduates.
  • Most investors didn’t find out that the government reimbursed CECO for both semesters at the start of the first, which should have meant no receivables.
    • This meant that those DSOs were quite high.
  • CECO was subsequently found to have enrolled students fraudulently.
• Example: Oracle
  • Oracle had a B/S line item called Unbilled Receivables.
    • When too many analysts questioned the item, ORCL aggregated it into gross receivables.
    • The stock collapsed a couple of quarters later amid channel-stuffing confessions.
• Another lucrative area of the B/S is reserves & depreciation.
  • Under-reserving, over-reserving, or reserve reversal to augment earnings “tell you a lot about the company.”
  • One fad-driven toy company had built up a huge reserve during good seasons because they expected to use it to insulate earnings when the fad burst.
    • However, the bad times lasted longer than the company expected and they ran out of the reserve, leading to several quarters of poor earnings.
• Although liabilities have become harder to follow—with most of the disclosure in the notes—items to watch include debt covenants, off-balance-sheet liabilities, and contingent liabilities.

2. Quality of earnings:

• • You can’t always tell what the business of a company is until you look at its financial statements.
• Example: Mail Boxes Etc. (ticker: MBE), a package-mailing store chain.
  • Their business was not sending packages—it was a franchisor, so the business was opening stores.
• Example: Marriott
  • MAR’s business is not running hotels—it’s opening hotels.
• You get a sense of a business’s model by looking at the gross margin and how it impacts operating income.
• Non-recurring earnings are important.
  • They can be hidden—for instance, as a gain on sale that offsets SG&A, interest expense, or depreciation.
    • For this reason, it’s important to compute line items on the income statement as percentages to detect radical changes.
• Example: Morrison-Knudsen (ticker: MRN) was supposed to be in the construction industry, but the company always had a lot of other income.
  • Other income was frequently gain on sale, interest income, and dividends on securities.
  • This lasted for about 3 years of the company’s vaunted turnaround under William Agee, until Morrison-Knudsen defaulted on its debt, to the board’s shock.
    • The base business had faltered badly, leaving them with no income to fall back on.
• Example: LA Gear once reversed unspent deferred advertising to pad earnings.
• Example: Dell Computer used “currency games” to increase earnings growth many years ago.

3. Quality of mgmt:

• • This used to be a lot easier than it is today.
    • One could look at mgmt’s income as a percentage of the company’s, and if it was 100%, they were overpaid.
    • Nowadays, healthy pay packages and golden parachutes are common and ”important”.
• Predominantly, you want to scrutinize self-dealing, past performance, and overpayment and options packages.
• Example: Crazy Eddie (ticker: CRZY) had massive related-party dealings, even funding a family-owned medical school with company money.
• Example: Stevie Hilbert of Conseco (ticker: CNC) married the stripper who jumped out of the cake at his son’s bachelor party.
  • Staley was short the stock on and off for 10 years and was “kicked out of many meetings.”
• The prospectus is a good place to find felonies, lawsuits, etc.
• Example: For one company (CMI—see below), which had no borrow but was a favorite of hers, the prospectus detailed that the CEO has been the head of a company called Cardio Pet (aka Animed, ticker: VETS), another one with no borrow.
  • Cardio Pet diagnosed animal heart conditions by taking credit card information over the phone and having the owner hold the receiver to the animal’s heart.
  • Note: she’s referring to Steven Rabinovici, who later ran Complete Management, Inc. (ticker: CMI), a Physician Practice Management company that listed in 1997 and declared bankruptcy in 1999 after numerous related-party dealings, heavily manipulated earnings, and a failed restructuring.
• Example: Walter Forbes, CEO of Cendant (ticker: CD), a company fined for accounting fraud in 2000, had done the same thing when the business was much smaller about 15 years prior.
  • The stock had gotten punished then, but everyone forgot, so he did it on a larger scale the next time.
  • Forbes had convinced investors that they weren’t smart enough to understand the financials.
• “Most human beings are one-trick ponies: we know how to do one thing and we do it over and over again.”
  • This is very true of CFOs: they will do it at every company they work for.

Case Study 1: Escala Group, Inc. (ticker: ESCL), a stamp auction company. Relevant 10-Q here.

• B/S:
  • Escala ran high inventories as a percentage of assets: 43%.
  • There is a small, but interesting, line on the September ’05 B/S for related party receivables, along with loan receivables and goodwill.
  • Current liabilities were way up—they had increased +230% QoQ and +190% YoY.
  • The key thing to find out: how did ESCL value inventories?
• I/S:
  • “Sales-trading” gross margins are extremely low, which you would expect for an auction business.
  • However, gross profit from “sales of inventory—related party” forms about 70% of overall gross profit.
    • The business of the company was the related party.
  • The related party was Afinsa, a Spanish company that guaranteed ESCL’s debt and owned the majority of the stock.
  • Sales-trading volumes were also almost $500 M in that quarter, which is very large, at least for stamps.
    • It turns out that ESCL’s annual volume was higher than that of the entire philatelic market.
• Subsequently, allegations surfaced that Escala had misappropriated funds invested by the Ohio Bureau of Workman’s Compensation.
  • Escala had marked up coins bought on behalf of the organization, but this investigation was not mentioned in financials.
• On June 5, 2006, Afinsa was found to have sold stamps to investors while promising a guaranteed annual return: Afinsa was a Ponzi scheme.
• ESCL’s inventory valuation turned out to be entirely fake.

Case Study 2: Novastar Financial, Inc. (ticker: NFI), a residential mortgage firm.

• Most old short-sellers were short Texas, Arizona, & New England real estate.
  • In the Arizona days, they would “fly out and count cranes” to know if it was time to short.
  • They would know each of the real estate lenders and what their mortgage portfolios looked like to be able to take advantage of popping bubbles when the time came.
• “When I started reading in the paper about the impending doom of the subprime market…it almost got me to go back to the business. Real estate is just so much fun to short.”
• Novastar originated, purchased, invested in, and serviced residential non-conforming loans.
  • It was also a REIT.
• In 2Q06 (10-Q here), mortgage loans held in portfolio had increased to $2.5 B, up from $29 M in 4Q05. The debt associated with such loans had also increased commensurately.
  • “My reaction was, maybe they’re just waiting for a pool to sell. Maybe it’s just a timing thing.”
• The I/S reflects the same thing: higher risk for not much more net interest income.
  • The provision for credit losses also grew +1200% from the start of the year.
    • “But, it may be because they’re holding more on the balance sheet.”
  • Gain on sale of mortgage assets was down -28% YoY.
    • “But, it could be because they’re carrying more and they haven’t sold it yet.”
  • Income from continuing operations was down YoY.
• So, the B/S & I/S raise some questions but not that many big concerns.
• Note 3 to the financial statements shows that the allowance for credit losses moved from 2.4% in 4Q05 to .4% in 2Q06.
• Note 5 discloses “other-than-temporary” impairments in NFI’s mortgages, leading to charges of $6.5 M vs. $1.7 M a year ago.
  • This begins to solidify the idea that 2Q06 is not suffering from a timing issue.
• Note 9 shows that NFI sold $793 M of mortgages in 1H06 with recourse for borrower default—against a reserve of $1.8 M—, and that they’ve bought back $9 M of mortgages.
  • That $790 M+ that can come back to bite the equity on the B/S later.
• The “Industry” section is very clear that mortgage spreads are narrower & that the housing market is shaky, with the subprime market looking a lot worse going forward.
• In 3Q06 (10-Q here), the financial statements show continued deterioration.
  • Interest expense from asset-backed bonds doubled sequentially.
  • The credit loss provision increased and net interest income fell, both sequentially & YoY.
  • In Note 3, the company introduced an allowance for doubtful accounts in its mortgage loans held for sale for the first time.
    • The allowance for credit losses for the held-in-portfolio mortgage loans also got a new, expanded table.
  • Note 9 discloses $1.5 B in recourse mortgage sales in 9M06, with reserves of $4.6 M. Mortgage repurchases in the period were $16 M.
  • The Industry section reveals that NFI was likely to fail REIT qualification for the year due to its high proportion of derivative income.

The current state of short-selling and other thoughts:

• Staley figured that Enron, WorldCom, etc. would have changed investors’ behavior.
  • This was not the case.
• “Most people seemed to think that fear was the only motivator in this world, and if we weren’t seeing the huge frauds that we did around 2000, it was only because the CPA firms, law firms, and investment bankers were afraid of jail—that they’d suddenly gotten religion and were paying more attention to the financials of the very big companies.”
  • There was general agreement among hedge funds she talked to that they weren’t seeing as many large-scale frauds.
    • Enron, Conseco, etc. had problems years and years before they blew up.
  • Fear of lawsuits has changed media reporting on malfeasance, too.
    • It’s harder for short-sellers to get the media to pick up earnings-manipulation stories by continually raising accounting questions before mgmt.
  • There is a “tacit conspiracy” between analysts, stockholders, and mgmt. Mgmts, which often just used to refuse to take questions after bad quarters, now have analysts who provide the patter to justify earnings performance as a one-off.
    • E.g., downplaying the impending effects of subprime.
• There’s also a liquidity problem: many funds don’t want to know if there’s a problem because they “have to” own the stocks.
• Negative rebates have also hurt shorting.
  • “Short-sellers are notoriously bad about timing: everything always takes 2 more years than you think it will.”
  • On the flipside, borrowing stock is now a profit center for brokerages, so there’s a lot more borrow than there used to be.
    • Short-sellers rarely get bought in anymore as a result.
• Most hedge funds aren’t doing fundamentally based short-selling.
  • They’re focused on negative earnings surprises, positive alpha, quantitative, arbitrage, etc.
• The prominence of private equity is nothing new:
  • “I lived through the days of LBOs, and if you’re a short-seller… you always have a buyout valuation in your head—high/low, based on anything you could.”
  • “One of the big risks in short-selling is that there are a lot of dumb companies around that will spend a lot of money on a bad company.”
  • The positive aspect of PE involvement is that they may monetize their investments at some point, which can be a good hunting ground.
• “Prospectuses are always good [hunting grounds]. IPOs are always great.”
• Hedge fund fee structures used to be very different: a 1% or 0% mgmt fee and a 20% incentive fee.
  • Most used to have a hefty percentage of GP money, so incentives were better-aligned.
• “Fund of fund guys are terrifically adroit at looking past the media, reputation, etc. of funds.”
• One study she read shows that 60% of short profits are made in 10% of the time.
• “Short-selling is a very risky business, and you usually don’t make any money doing it at all. It’s very company-specific, and psychology, liquidity—which is the market—are the primary movers of stock price.”
• “Most Wall Street analysts really can’t do accounting.”