Q: Your fund’s inception date is August 2000. What is the history of the portfolio before that? A: I was recruited by Vanguard in 1997 from Merrill Lynch Trust Company to head up Vanguard’s Asset Management and Trust Group. At Vanguard, we received a lot of assets from high-net-worth individuals that were individual securities and we couldn’t just sell those securities and put them into Vanguard funds – that was clearly not the proper or right thing to do as a fiduciary. We had to develop a way within the Vanguard Trust Group to analyze individual securities and with a couple of CFAs and other trust and asset-management professionals we built a quantitative model that was a hold-and-sell discipline at the time. When we inherited an individual security, we had to determine whether that particular security statistically had a probability of actually underperforming its peer group or the market by greater than 20%. If it dIdent, then we could sell it because of the capital gains consideration. Initially, we would look at between 25 and 30 factors to determine whether there was any correlation or predictability correlation amongst these factors. Then we went to Wellington and had it back-tested, then we back-tested it amongst ourselves and came up with a pretty good model that could tell us if we inherited, for instance, Wells Fargo, whether or not we should hold it or sell it. And as I got to know and see how our model really worked well, I looked at it and said: this is not just a hold-sell model, this really is a robust quantitative model that could be buy-hold-and-sell, and we could actually run money by it. So, when I decided that I really wanted to start my own firm and left Vanguard, I bought into a small private bank called Millennium Bank. And Millennium Bank was founded as a private bank by the former chairman of Meridian, the former chairman of PNC, and the former vice-chairman of Lehman Brothers. It was a good group of progressive bankers and they had a small mutual fund company and I ended up heading up that group as a president of the mutual fund, called the Penn Street Fund. It only had one portfolio at the time, which was a balanced fund. And I said: “Guys, I have a tremendous model that we ran and developed in the past that we could rebuild and use a couple of different factors to turn it into a hold-buy-sell or a real quantitative model for the buying and selling securities.” So, we did that. We invested our own and the bank client’s money into this portfolio and we said: “Well, rather than go out and pre-sell what we think that we have built, let us run it for three years with actual money, with my own money, with some of my clients’ money in this, and show the world, and demonstrate to everybody that what we have built actually works.” So, we didn’t go out and try to bring in a lot of assets at the time, we ran the money – about $4 million – for three years and we continued to show outperformance. And now, with the three-year number hit in August 2003, what we said we would do, we actually accomplished, and even better than we thought we would. Q: Yes, that model proved itself in that particular environment during the past three years, but what else, in your view, helped it beat the market? A: I think one of the reasons that we were able to achieve the results that we have achieved is the process that we put in place. The selection of individual securities, although important, was not the driver of our performance. I think the driver of our performance has been the actual selection weightings, overweightings or underweightings of particular sectors. This is what makes us so unique. We look at the top 1250 securities in the country by capitalization. We run them through the model. The model itself has about 33% growth factors, 33% value factors, and 33% momentum. So, rather than selecting a particular style – growth, value, or momentum – we look at all three, and therefore we get proper diversification on the downside. After we run the 1250 securities, we look at the top 125. And then, by pure count, how many stocks fall into the eleven sectors that we have identified – like consumer cyclicals, to consumer staples, to healthcare, to financials, to technology – we set our target weighting on those particular sectors. Q: And what are some of the factors that determine the total rank of each particular stock? A: There are six factors that we look at, and all of those factors have additional sub-factors that drive them, but absolute price-to-earnings ratio, absolute price-to-book ratio, volatility of adjusted earnings surprises, time-weighted earnings revisions are among the top ones. Q: And all those things have different coefficients and they decide the final score? A: That is absolutely correct. We have the growth ability index and the 9-month price trend. And that rounds up our six major factors that go into the model. Q: What happens after you have your weightings determined? A: Now, that we have our weightings determined, it comes to the security selection. What we will do is we will decide, let’s say, that we participate in ten of the 11 sectors, or nine of the 11 sectors, but we will always be participating in six. We can never overweight a sector more than 200% of the Russell 1000 weighting. We never make unbelievable holistic sector bets. So, I am rotating in and out of sectors relative to the Russell 1000 weighting. It gives a downside protection, it is a risk control. We will decide what the actual sector weightings are going to be, we make sure which ones we are going to participate in completely, and then we will do security selection. Now, security selection may not be just from the 125, but from the top 250. I will hold between 50 and 70 securities at any given time. Q: Did those sector weightings change somehow over the past three years? A: If I would have predicted how the model will operate in sector weightings if this was five years ago, I would not have predicted it. And that is a nice part about it – the human prediction factor gets thrown out the window and the model says to me: “Here’s what you have to do, Mike.” We are seeing a shift more in the style rotation – growth and value. The model had me drift towards value from the inception in 2000 probably to the end of 2002. Not exclusively value, but it had me overweight value to the extent that if you look at it, you would say: “Geez, this is not a multi-cap core, this is a multi-cap value with some growth.” But that is what the model told me to do. Since the beginning of 2002, the composition of the portfolio has become more growthy, and therefore I truly have come back towards growth and truly am what we always said we would be – a multi-cap core right now. We don’t do it internally, I’d rather have an outside company like Lipper tell me if I am growthy, or I am value-oriented. Again, it is not my decision, which I like, it is the model telling me what to buy and that is the process. We underweighted technology significantly and the model told me to do that. The model has told me to significantly overweight financials. We had more stocks within the top 125 to fall into the financial industry than into any other industry. And some of those over the last three years that I have seen fall into the banking segment of the financial industry. It hasn’t consistently always been banking, but it has consistently always been financials. Q: So, the signal was to be underweighted in technology, overweighted in financials, and what else? A: And to be completely out of communication services. Q: What about the last year? A: In the last year, what I see changing is that I see my weighting in technology is increasing. The model told me to increase it. Not to the point where we were market neutral, we are still underweighted relative to the market, but relative to where we were 12 months ago, we have some more stocks in the top 125 than we did a couple of months ago. Q: What is your turnover? A: The turnover for this portfolio is impossible to say what it will be ten years from now, but I can tell you about the last three years: it was not what you would expect from any long-term investment portfolio. It has been about 70% on a yearly basis, between 50% and 70%. Q: Do you have any limitations or restrictions in terms of what securities you should buy and not buy? A: We are free, provided that if falls within the top 1250 U.S. domestic equities by capitalization. Q: And which companies make the cut right now – mid-cap and large-cap only? A: This would be called a multi-cap core. It is mid-cap and large-cap, and not all mid-caps – from the mid-level of mid-cap and above. We rely on data coming in from other sources and when you have to do some fundamental analysis, it is difficult to rely on just pure data at the small-cap level. Q: Since you don’t visit the actual companies, who do you rely on supplying the relevant data? A: I don’t do fundamental analysis, but one of the factors is that I rely on what the analysts are saying out there. They are doing fundamental analysis. So, if a company had no analyst following it on a national level, it is not going to qualify for the model – it is going to rank really low, because it will score zero on most factors. So, to have a national-level analyst, or a number of national-level analysts following it, is important. Otherwise, it may go to the top 1250, but it is going to score horribly in the model, because it has to have some analyst following. Q: Have you ever had to make tweaks and adjustments to the model itself? A: At this moment we always look at the different factors, we run regressions and all that kind of stuff, but we have at this time in the last five years that I have looked at the model, we have not made any adjustments in the actual weightings of the factors that we look at in the model. Q: As a portfolio manager, you are also obliged to tell the people what can go wrong with the model? What do you think can break it? A: “What goes wrong” is first and foremost in my mind, because I come from a trust environment. I have to protect the principal. What could go wrong? Let’s say you have a particular bounce in the market on an upward basis, it doesn’t make sense according to the factors that we have in the quant model. The market gets ahead of itself. Take technology. For some reason, psychologically, it has drastic upswings. We are going to miss that. What we are seeing is, when that does occur, there is a regression back to where it should be. And when the actual earnings come out, that is when I get a lot of outperformance. But when a particular sector like technology takes off like crazy, I am not participating. The model won’t let me do it. I have to stick to the discipline in the process. What I would do is I really believe that I trade off the fundamental or the psychological variations in the market for process and speed, without anybody, Mike Mara, or any other portfolio manager saying: “God! I love this particular stock right now. It doesn’t rank well, but I am going to buy it on a hunch, anyways.” I can’t do that. Q: What would you tell an investor who is considering to put some money into your fund? What can they expect and what they should not expect? A: From the Sector Rotational portfolio, an investor can expect a consistent process in which you are participating in all styles in the market at any given time, and participating in the majority of industry sectors of the market at any given time, which is fully invested and does not make bets as to whether the market is going to be up and down this month or next month. This model, this portfolio is for the long-term investor. It is not one which is designed to help the market timers. It is not good for them, it is not attractive to a market timer. If I am an investor, it would be a wonderful place to build my core, and then, if I want to make bets, let me buy a small-cap growth company, or let me buy something that is out there on the Richter scale. But for the core, it is a great alternative for the S&P 500. It could be in anybody’s portfolio. What they should not expect is in up markets, driven by one particular sector, we will not shoot for the moon. We won’t participate in any single sector that will take the market away, because we fundamentally believe that there is always a reversion to the mean. We just believe that our mean will always be higher than the average market itself as established by the Russell 1000 or the S&P 500.