Mohnish Pabrai on Mistakes & Learning

Page 12 what to do with it sometime between 2025 and 2035. Seritage is in the funeral and recycling business. I wrote in my first book Mosaic why funeral businesses are such awesome assets to own. Micron is in an industry with rapid change but it is also in a stable oligopoly with a solid moat. The business has gotten a lot better since we invested in 2018. I continue to watch it carefully. Mumbai real-estate has finally woken up from its long slumber and Sunteck has many tail winds. The plan is to hold it for a while. The beauty of this ownership mindset is that it can tolerate a healthy error rate. We do not need to be right on all five bets. We may end up with great results even if just one or two of these bets work out as long as the others are flat. The odds of permanent loss of capital are very muted in all five bets and we could be very right on at least three out of five of them. I am always reticent about discussing current portfolio positions. It causes commitment and consistency biases which can hurt us. There are no guaranteed winners. Not even Reysas. I am hopeful that the portfolio continues to be managed objectively and rationally without bias. All three funds have similar, but not identical holdings. Most of our Reysas shares are owned by PIF3. Reysas makes up about 8% and 2% of PIF2 and PIF4 assets respectively. PIF2 and PIF4 have some great holdings that aren’t present or meaningful in the PIF3 portfolio.

mistake randomness for a set of determinate patterns. Well worth reading, even for a value investor. Assets Under Management There was about $59 Million in assets under management between all the funds as of October 1, 2003. Thanks for your continued interest, referrals and support. Pabrai Investment Funds Assets Under Management (In Millions of $) 1999 2000 2001 2002 2003 Assets Under Management Feel free to call me at +1949.275.5652 or email me at mpabrai@pabraifunds.com with any queries or comments. Warm Regards, Mohnish Pabrai Page 7 of 7

The annualized performance % numbers for The Pabrai Funds were off as I had been using an errored formula. There is no error in the NAV/unit numbers. Thankfully, the mistake understated annualized performance. Our actual annualized performance is better than previously reflected. This has been corrected. I’m sorry about this. The diligent reader will notice that PIF2 had a strong December and a very good 2001. I had not published our 1/1/01 NAV before, but it was $9.87/unit. For the year 2001, our gain was 57.1% before fees and expenses. Considering that all three indices were down, we had an exceptional year – one that is clearly unsustainable. When compared to 2001 mutual fund performance, PIF2 would have come in as the 5th best performing fund of the year including ALL mutual funds. Of the 4 funds ahead of us, 2 are country specific funds (Russia and South Korea). The data indicates that we’ve outperformed over 99.9% of mutual funds in 2001. As I’ve stated earlier, I think we’re fairly good, but not that good. Long term, I’d expect more funds to outperform The Pabrai Funds while we continue to outperform the indices and atleast 90% of fund managers. We achieved these returns with modest realized gains in 2001. Our portfolio continues to trade substantially below intrinsic. I’m not sure when Mr. Market will assign an appropriate value to the holdings, but I have some confidence that, for most positions, this will happen within 24 months. So, we have a few “Aces in the Hole.

The name was maintained (as these small funeral homes had tremendous brand equity in their communities), but the back-end, merchandizing, selling pre-paid funerals etc. was streamlined and corporatized. The mistake all three made was that most of the acquisitions were done for cash rather than stock and they freely borrowed money to support their acquisition binge. Then the music stopped. The street lost its excitement for the funeral business and their stocks started to come down. Bankers and lenders weren’t interested in further lending and wanted balance sheets deleveraged. With no additional acquisitions, sales went flat. This is not a growth business. Stewart found itself in 2000 with $930 Million of long-term debt with about $500 Million coming due in 2002. The company had said to the street that they had excellent relations with their bankers, had never violated a bank covenant and were confident that their bankers would extend the loan. They also said that they had begun to explore the sale of international funeral homes and cemeteries in Europe, Mexico etc. International assets comprised about 20% of revenues and assets, but weren’t generating much cash flow. Thus Stewart has about $460 Million in assets outside the US. Despite this data, the stock was at $2/share. I figured that there were two possible scenarios – the company would either be forced into bankruptcy by lenders or they would refinance/eliminate debt and continue as usual.6

Page 3 7/1/99 – 6/30/00: 150% 7/1/00 – 6/30/01: -30% 7/1/01 – 6/30/02: 200% 7/1/02 – 6/30/03: -35% Before fees or penalties, if there were $1,000,000 under management on 7/1/99, it would have grown to $3,412,500 or an average annual return of about 36%. Dalal Street’s fees/penalties would have been: Start Pre-fee End Post-Fee End Fee 7/1/99 – 6/30/00: 150% $1M $2.5M $2.05M $350K 7/1/00 – 6/30/01: -30% $2.05M $1.44M $2.17M ($610K) 7/1/01 – 6/30/02: 200% $2.17M $6.51M $5.43M $1090K 7/1/02 – 6/30/03: -35% $5.43M $3.53M $5.76M ($2.23M) In other words, Dalal Street is in the hole by $1.4M and investors endup with an after fee return of $4.76M after 4 years or an average annualized return of about 55%. There is something drastically wrong with this picture. Buffett avoided the Case 1 problem completely by setting up new partnerships whenever he got new funds. I don’t like that solution to the Case 1 problem because I’ll end up managing 20 very small buckets of money. Very inefficient and time consuming with lots of record keeping and investing overhead. He avoided Case 3 by never having a down year. He never had a year where returns were over 59% or less than 6.8%. I’ll exceed the 59% record most likely in the first year. It is very likely that if first year returns are north of 100%, the next year may be negative or marginal. Therefore, I need to alter PIFI rules to fix the aforementioned problem. Here is the proposed solution: 1.

Funds like to categorize themselves as “growth” or “value” or “small cap” etc. I find that with many stocks, “growth” and “value” are two sides of the same coin. They are not mutually exclusive. I get a lot of questions from investors regarding the inner-workings of this “black-box”. I’m including some very specific information on The Pabrai Investment Funds to shed some light in this matter. The investment style of the Pabrai Investment Funds is quite simple. The fund only takes long positions in public equities. There are no options or derivatives etc. that the fund delves into at all. Typically the funds assets are divided between under 15 securities with the typical allocation for a given security being 10% of assets in the fund. The fund is allowed to go up to 30% into margin. I look at 3000+ public companies a year. When I look at a given public company, I run them through a three-question filter. 95+% of companies do not make it through these filters and are discarded. The ones that make it through are then rigorously analyzed before anything becomes part of the portfolio. Inspite of my best efforts, I have made mistakes in the past and know that there will be more in the future. The goal is that we are right many more times than we are wrong. The three filters that a security has to go through is a positive answer to the questions: 1. Do I understand this business well? Is it well within my circle of competence?