Fooling Some of the People All of the Time, a Long Short (And Now Complete) Story, Updated With New Epilogue (5 page)

It turned out that raising $10 million to start Greenlight proved too ambitious. As we went through the contact list and took whatever meetings we could get, we soon realized that almost no one would invest with a couple of twenty-seven-year-olds with no track record. We decided that the only way to get a track record would be to get started. In one year we could have a one-year record, and in three years a three-year record. It was not going to happen any faster than that.

We launched in May 1996 with $900,000—more than half from my parents. Our initial investments included MDC Holdings, a homebuilder that we still own, and EMCOR, an electrical and mechanical systems contractor that had recently emerged from bankruptcy. We made a good profit on EMCOR, though it took until 2001 before it really worked.

We made 3.1 percent in May 1996. (Whenever I cite Greenlight fund returns, they are after fees and expenses, that is, the “net” to the partners unless otherwise indicated as “gross.” I always discuss the impact of individual investments on the gross return.) At the end of the month, we invested 15 percent of the fund in C. R. Anthony, a small retailer that had recently emerged from bankruptcy and returned to profitability. The market valued the company at $18 million despite its having twice that in net working capital (current assets less all liabilities). Greenlight returned 6.9 percent in June.

In July, the market suffered a correction and the S&P 500 fell 4.5 percent. However, our portfolio enjoyed several good events and generated a 4.8 percent profit. We had bought bonds in the campsite operator U.S. Trails at 77 percent in June, and the bonds got called at 100 percent in July. We made a nice gain when the semiconductor capital equipment manufacturer Tylan General announced it would be sold at a good premium. Finally, our larger short position (we had only two at the time), Microwarehouse, announced terrible results due to systems problems, and the stock collapsed.

After the close of trading on the last day of each month, I stood at the fax machine and sent the statements to the partners one at a time. Most of the people we met before we launched asked to be kept informed, whether they meant it or not. Now, a few began to notice and send money. We got our first million-dollar partner that August.

The year could have gone better only if we had not missed some opportunities. At one point during the summer, I considered an investment in the creditor claims in the bankrupt retailer Best Products. I finished the work, but rather than buy the position, I decided to “sleep on it.” I came in the next morning and told Keswin that I wanted to make it a 12 percent position. I called the salesman I had discussed the idea with at a brokerage firm that traded the claims and gave him my order. He asked if I had seen the news. I hadn’t. Service Merchandise had agreed to buy the company, and the claims had doubled overnight. Of course, making a mistake on an actual investment is far direr than missing a good opportunity.

Another of our initial partners thanked us for the good results by giving us a list of about a dozen of his “wealthy” friends. Most became partners. One did not invest because he presented me with a brainteaser card puzzle that I couldn’t solve. He asked me to take one suit from the deck and arrange it so the cards would appear in sequential order when I turned the top card face up, put the next card on the bottom of the deck, turned the next card face up, put the next card on the bottom of the deck, and so forth. I blew it: I would have to work on my card skills. (The correct order is A, Q, 2, 8, 3, J, 4, 9, 5, K, 6, 10, 7.)

Greenlight returned 37.1 percent in the last two-thirds of 1996 without a down month. Our assets under management hit $13 million. We decided to have a “partners’ dinner” to explain our results and rented a small room in an Italian restaurant on the Upper East Side of Manhattan. The partners came on a snowy evening in January. And they weren’t just Mom and Dad, but about twenty-five people—almost everyone we invited, including several from outside of New York. We gave a presentation of the business and the results. It was not hard, as both longs and shorts contributed, and we did not have a significant money-losing investment to discuss. The results were led by C. R. Anthony, which had increased 500 percent and generated about one-third of the return.

The next day, Bruce Gutkin, one of our four “Day One” investors, who would eventually become our head trader in 2004—originally our only trader, but this is an age of title inflation—called to say not only how enjoyable the dinner was, but that his wife remarked on the way home that “this is how big things get started.”

CHAPTER 3

Greenlight’s Early Successes

We started 1997 strong and returned 13.1 percent in the first quarter. Then, I made my first costly mistake. There are two types of bad outcomes. Sometimes, after analyzing the risk and reward, an investment appears attractive, but the unfortunate or unlikely happens. Such is life. Other times, the analysis is simply flawed: The investment is poor and we deserve the eventual loss. This mistake was the latter. We invested 6 percent of capital into Reliance Acceptance, which charged 18 percent for car loans to people with tarnished credit. The key investment issue: Was that 18 percent enough to cover the losses from loan defaults, which were harder to estimate? I analyzed the car repossession data and determined that Reliance repossessed 20 percent of the cars and lost 40 percent of the loan each time. The loans lasted two years, so I calculated annual losses to be 20 percent × 40 percent ÷ 2, or 4 percent. The high interest rate appeared sufficient to cover the losses and the stock appeared cheap, at a discount to book value.

I erred by not framing the loss analysis properly. The repossession statistics did not include about 10 percent of the loans where the repossessor
could not find the car
. Obviously, these loans were 100 percent losses. This meant the real losses were more than twice what I’d calculated. The 18 percent interest did not cover the cost of funds, the true losses, and the operating expenses. We lost about half our investment before I realized my error. This led to our first down month in April, where we lost 0.3 percent.

The rest of the year was a cakewalk. The biggest winners: insurance company demutualizations, spin-offs, Pinnacle Systems, and some short sales. Demutualizations are good hunting grounds for our type of investing. Many insurance companies have been formed as customer cooperatives, or “mutuals.” In a mutual, there is no share ownership, but policyholders, who are considered the “owners,” do have some rights, such as electing directors. Management’s simple self-interest is to stay solvent. They tend to have conservative accounting policies because there is no stock price or even an organized ownership group to worry about. On the margin, large reported profits generate taxes and raise the possibility that the policyholders might demand some of the money back, either through lower premiums or surplus payments. Reporting profits actually could lead to eventual financial trouble—the one thing management needs to avoid.

From time to time, mutuals convert themselves to stock companies in a transaction called a demutualization. The most attractive deals are 100 percent sales of the stock in an initial public offering (IPO), with the proceeds going to the company. The new investors effectively get the company for free, as ownership of the post-IPO stock includes both the IPO proceeds held at the company and the company itself. Add in a nice dose of stock options for management set at the IPO price, and the incentives and the structure allow new shareholders and management to make a killing with little risk. In many cases, lackluster-performing companies begin to show remarkable profit improvements after the IPO.

Some spin-offs have similar dynamics, though they need to be assessed case by case. A spin-off is when a large company divests a subsidiary by distributing the subsidiary’s shares to the parent company’s shareholders. Over the years, we have found that carefully selected spin-offs are terrific opportunities.

Pinnacle Systems was a technology company that had reported a couple of disappointing quarters. Its stock traded down to book value, which was mostly cash. Many value investors eschew investing in technology companies because the products are complicated and the field changes rapidly. We take the view that technology companies that are not losing money, are trading at book value, and appear to have a viable product are good investments. It proved to be the case in this instance, and when Pinnacle reported better results, the shares tripled.

Finally, some of our short sales made nice contributions in 1997, including Boston Chicken and Samsonite. Boston Chicken’s accounting practices enabled it to recognize up-front revenue and profit when franchisees opened restaurants. Boston Chicken financed the openings and up-front fees and earned interest on loans to the franchisees. The underlying restaurants were not profitable enough to support the payments to the parent. Boston Chicken’s shareholders were not concerned, or perhaps were not even aware that franchisees lost money, because Boston Chicken did not consolidate the franchise operations in its financial statements. We believed that if the restaurant economics were not robust enough for the franchisees to satisfy their obligations to Boston Chicken and make a reasonable return for themselves, they would stop opening more restaurants and Boston Chicken’s price-to-earnings multiple would fall as it stopped growing. It turned out even worse because the franchisees defaulted on the loans. Eventually, Boston Chicken went bankrupt.

Samsonite also collapsed. We sold its shares short at $28 and watched them soar to $45. I checked and rechecked the thesis and decided to suck it up. Samsonite had raised prices and broadened its distribution network at the same time. It had opened many of its own stores, which aggressively competed against its own wholesale customers, the retailers. We saw a luggage store in Manhattan with a window display sign promoting “Samsonite 40% off.” We bought the sign to hang in our office. The clerk gave us a funny look. It turned out that wasn’t the only store working off excess Samsonite inventory. When Samsonite acknowledged that consumers didn’t accept the price increase and retailers were awash in excess inventory, the shares collapsed to $6.

We hired our first employees in the summer and moved into our own office in the Graybar Building, next to Grand Central Terminal. Our 1,300 square feet felt like a palace. I had my own office and could talk to my wife on the telephone without anyone knowing what we would be having for dinner that night.

We ended 1997 up 57.9 percent with $75 million under management. We decided not to accept additional money until we were prepared to invest it. Why? Adding too much new capital to a portfolio too quickly is a problem. It creates undue pressure to find new investments or to add to existing positions. We do not deploy new capital into existing positions unless they are either fresh ideas or positions to which we really want to add. However, while professional money managers habitually put new money into existing ideas, we don’t feel comfortable doing that when an investment is already in the middle innings. If we buy something at $10 thinking it is worth $20, do we really want to add to it at $16 if we think the value hasn’t changed? It is better to wait for a fresh opportunity or to close the fund to new investment. On the other hand, when the portfolio is fully invested, adding new assets helps investment performance. It allows room for new opportunities without selling existing positions prematurely. We have accepted new capital from time to time under those circumstances.

Fifty people attended the 1997 partners’ dinner at the Penn Club. We expected it to be a celebration. It was not. After our presentation, we took questions. Several partners complained about how fast the assets under management had grown. They worried we would not be able to keep up the returns. I explained that we closed the fund and would not accept new money until we were fully invested and emphasized that we did not expect to make 57 percent ever again, under any circumstance. As we never expected the results to be so good, we did not believe them sustainable at any asset size. Our goal is to make 20 percent per year. This will not happen each year, but we hope to average that over time with demonstrably less risk than the market. This is a challenging goal, which we may not achieve. I believe in setting high goals rather than easily clearable low ones. However, the strong initial result was no reason to raise the bar. No matter, the dinner was a tough experience. I learned that if we were going to have question-and-answer sessions, I had to be prepared for anything.

We started 1998 well, as the fund returned 9.9 percent through April. Then, Computer Learning Centers (CLCX), our largest short position, became a problem. CLCX was a for-profit education company that took advantage of generous government student loans and ripped off both students and the government. The company charged $20,000 a year to teach computer skills to uneducated people on obsolete technology. They accepted anyone. Another short-seller sent someone to intentionally fail the admissions test at one of the schools. The admissions officer gave her the answer key and then asked her to take the test again. Because the company offered a poor product and engaged in misconduct, we took a large short position. A local TV station in Washington, D.C., ran a feature that interviewed a bunch of angry students complaining about the poor facilities and showed an admissions officer on hidden camera promising a prospective student an absurdly high expected starting salary upon graduation. The stock market reaction: yawn.

CLCX announced a strong first quarter. Reid Bechtle, the CEO, confronted the short-sellers, telling
The Washington Post,
“Every dollar the stock goes up is $4 million the shorts take out of their bank accounts.” The
Post
said he told an investor, “We’ve already gone through Hiroshima and it’s time for Nagasaki” for the shorts. Shortly thereafter, the shares began to decline when the Department of Education announced a program review to examine compliance and the Illinois attorney general filed a civil fraud complaint. The stock sank. Sensing that the end was near, we increased our short position.

A couple of months later, CLCX paid a $500,000 fine and promised better business practices to settle with the Illinois attorney general. The attorney general thought this was a big penalty, but the stock market judged it a trivial cost of putting their problems behind them. Bulls spread the word that the Department of Education completed its program reviews and would not take strong action. Three large mutual funds in Boston each added to already enormous positions. The stock doubled quickly. It looked as though CLCX might actually get away with it. I decided to swallow my medicine and covered the short in July. We lost about 2.5 percent of our capital on that position.

Covering the short was a poor decision because it turned out we were right about the company. The publicity from the regulatory action and more conservative behavior by the company caused enrollments and earnings to fall short of expectations, which killed the stock. This actually happens a lot in controversial short sales: Many times, the bulls win the battle on the core criticism (in this case, regulators didn’t immediately kill the company), but the bears win the war, as business or accounting reforms cause disappointing performance. It took the Department of Education two more years to complete its work. When it did, it demanded that CLCX return all the student loans it had ever advanced to the government. This put them out of business. (For a good summary, go to
http://chronicle.com/free/v47/i23/23a03501.htm
.)

A key problem for investors who short a company that is subject to government oversight is that the government, even when it acts, does not move at the same speed as the stock market. Two years might make a prompt government investigation, but it is an eternity for investors such as Greenlight reporting monthly results, even in a long-term strategy. Based on my decision to cover CLCX, I developed a stronger stomach and learned to become even more patient. CLCX is one of the more expensive of many examples that have taught me this lesson.

Unfortunately, as we covered CLCX, the stock market made a near-term top. Around that time, we also covered a couple of other successful shorts. One was Sirrom Capital, named for the founder, but with his surname spelled backwards. Sirrom was in the same business as Allied Capital, with which we were unfamiliar at the time. Sirrom was a business development company (BDC) making mezzanine loans (senior to equity, but subordinate to senior debt) to private companies.

BDCs are a special creation, formed by Congress as a way for small businesses to have more access to capital and receive professional management expertise. They have existed in some form since the Investment Company Act of 1940, but their current structure was born through Congress with the Small Business Investment Incentive Act of 1980. BDCs lend small businesses money, advise them, and in return collect interest and fees. In essence, BDCs are publicly traded private-equity firms that give the public an opportunity to participate in the growth of young companies. BDCs raise capital in equity offerings and act like closed-end mutual funds. They are subject to the Sarbanes-Oxley Act of 2002 and the Securities Exchange Act of 1934. BDCs are required to maintain 200 percent asset coverage on the debt they issue. In other words, the value of the assets they invest in must be twice the amount of debt they take on, which caps their ability to leverage. They also don’t pay corporate taxes, provided they pass through their taxable earnings directly to shareholders.

Sirrom funded rapid growth through a virtuous cycle where it raised equity at a sizable premium to net asset value (book value), which increased its net asset value and provided fresh, cheap capital to grow its portfolio. This cycle enabled Sirrom to grow its earnings and dividends, which caused the stock price to appreciate further, allowing Sirrom to repeat the cycle beginning with another stock offering.