Favorite Links

Sunday, July 1, 2018

(CHOIVO CAPITAL) To Diversify or Not to Diversify, That is the question.

Author: Jon Choivo | Publish date: Sun, 1 Jul 2018, 12:42 AM

To Diversify or not to diversify, that is the question. And that was a question i've constantly asked myself. The following few pages from "The Art Of Value Investing" by Whitney Tilson really helped me. I hope you will find it useful as well.


One of the most basic elements of portfolio strategy is determining the number of positions to hold. While the subject of much analytical research over the years—about which accomplished investors are typically well versed— the question of how concentrated or diversified one's portfolio is often appears driven as much by personal experience, comfort level, and “feel” than anything else.

* * *

When Warren lectures at business schools, he says, “I could improve your ultimate financial welfare by giving you a ticket with only 20 slots in it so that you had twenty punches—representing all the investments that you got to make in a lifetime. And once you'd punched through the card, you couldn't make any more investments at all.” He says, “Under those rules, you'd think carefully about what you did, and you'd be forced to load up on what you'd really thought about. So you'd do so much better.” It's not given to human beings to have such talent that they can just know everything about everything all the time. But it is given to human beings who work hard at it—who look and sift the world for a mispriced bet—that they can occasionally find one. And the wise ones bet heavily when the world offers them that opportunity. They bet big when they have the odds. And the rest of the time, they don't. It's just that simple.

—Charlie Munger, Poor Charlie's Almanack

The strategy we've adopted precludes our following standard diversification dogma. Many pundits would therefore say the strategy must be riskier than that employed by more conventional investors. We disagree. We believe that a policy of portfolio concentration may well decrease risk if it raises, as it should, both the intensity with which an investor thinks about a business and the comfort level he must feel with its economic characteristics before buying into it.

—Warren Buffett, 1993 Berkshire Hathaway Shareholder Letter

Value investors should concentrate their holdings in their best ideas; if you can tell a good investment from a bad one, you can also distinguish a great one from a good one.

—Seth Klarman, The Baupost Group

In our separate accounts we typically hold 10 to 15 securities. In our partnership we're typically more concentrated, with the top five positions making up about 65 percent of the portfolio. If I didn't have partners, our concentration would be even higher. You know how much of Warren Buffett's partnership held in American Express when he bought it after the DeAngelis salad-oil scandal? 40 percent or so. A company compounding capital at way above-average rates, when I have great confidence that will continue and the valuation is modest, I want to hold it at a size where it can have a material impact on the portfolio. The rationale is that simple.

—Chuck Akre, Akre Capital Management

We typically have 15 to 20 positions. Our general feeling is that if we don't like something enough to own a 5 percent position in it, we should wait to find something else. Put another way, if the only way you can feel comfortable about an idea is to own less of it, to my mind that tells you something about the quality of the idea.

—Andrew Jones, North Star Partners

We're deliberately concentrated on 10 to 14 investments, for two reasons related to time. First, it takes considerable time to learn enough about a company, its people, and its industry to develop and maintain a proprietary level of insight information, or knowing more than the Street. The second relates to our focus on activism: pushing for change at companies takes a lot of time. I will say that I have in the past fallen into what I call time traps, where I've spent too much time trying to resolve problem investments. We will pick our battles, but usually we're better off helping our best investments maximize opportunities than trying to perform brain surgery on dogs.

—David Nierenberg, D3 Family Funds

Owning fewer than 15 stocks I'd have more risk of being wrong with one company than I'd like. At the same time, if I owned 30 to 40 stocks, experience tells me that roughly half would be my favorites and other half would either have less upside potential or more risk. Rather than force myself to make that choice between less upside and more risk, I would rather just limit the number of individual holdings to 15 to 20. Most managers would buy the stocks with more risk rather than give away upside, which can often offset whatever benefit they think they are getting from greater diversification.

—Ed Wachenheim, Greenhaven Associates

We're not playing a probability game, where you invest in 100 businesses and do fine if 70 of them succeed and 30 do poorly. We're trying to select 25 to 30 businesses to own and we count on all of them doing well over a five-year period.

—Timothy Hartch, Brown Brothers Harriman

We think there are profound research advantages in concentration. This is a world of very smart people and you can't enter that marketplace without a profoundly humble view about how you're going to win. For us, concentration and depth of research allows us to go to bed at night and feel like we have a defensible source of our returns.

—Adam Weiss, Scout Capital

We ideally hold 25 to 35 positions, with a core holding at around 5 percent. That gives us most of the free lunch of diversification and allows us to maximize return by owning only our best ideas. There's also a human element to limiting our number of positions. With 25 to 35 stocks, our entire investment team knows every company and can have a clear opinion on it. With 100 stocks, you can't do that. The portfolio manager can only know how well each individual analyst is doing by the performance of his or her picks, as opposed to evaluating the decision inputs. Suddenly, the reality of reviews and compensation force you to look at that performance over shorter time periods than you should. The analyst knows that, of course, so then starts worrying about whether Wal- Mart's same-store sales next month are going to disappoint, rather than whether the company is creating long-term value. So we think a long-term strategy just works best with no more than 35 stocks.

—Boykin Curry, Eagle Capital

We typically hold 30 to 35 stocks. We cap any one position at no more than 6 percent of the portfolio, but we won't put anything in the portfolio at less than a 2 percent position. Setting things up this way keeps us from being distracted, makes us dig that much harder for truly interesting ideas, and forces us to make active decisions. Any one holding is too important to let slide if it's not working, and there's always healthy pressure on existing holdings from new ideas. We want to take away the drag of inertia, which can be very strong in human psychology.

—Mariko Gordon, Daruma Capital Management

We're fairly concentrated, with about 70 to 75 percent of our capital in our top 20 positions, so we know what we own and don't need a lot of statistical analysis to figure out where we're exposed. We think concentration is the key to big performance, but we also have no desire to have our year depend on one or two things working out, so we have generally kept our largest positions at 5 to 8 percent of total capital and make sure those big positions are not particularly speculative or highly levered.

—Gary Claar, JANA Partners

Our view is that too much diversification in many cases reflects the fact that the portfolio manager isn't doing the work to fully understand the businesses. If you do the work and find a great business run by great managers at a great price, that ought to express itself in the size of the holding within your portfolio. It's uncommon for us to go above 10 percent on cost in a given name, but we've held appreciated positions as high as 20 percent of the portfolio. We generally don't think you can have too much of a good thing.

—Peter Keefe, Avenir Corp.

When we see competitors holding 75 to 100 positions, with 75 to 100 percent annual turnover, we're either very impressed with their ability to keep track of 150 or more companies . . . or we're skeptical of their ability to credibly follow that many companies. We don't have anywhere near that many good ideas in a year.

—Eric Ende, First Pacific Advisors

Given that our funds are concentrated both in the absolute number of positions we hold and in the number of industries that are represented, it's natural for our performance to be lumpy. If the alternative is being consistent but mediocre, we would much prefer to be streaky but good.

—Wally Weitz, Weitz Funds

Diversification is an important part of our risk management. Average individual positions range from 1 to 2 percent, with the largest core positions at 4 to 5 percent. In 15 years, we've had three positions that got as high as 8 percent, two that worked out very well and one, Tyco, that was a disaster at the time. An important percentage of the firm's total capital is our own money and we're just trying to do what we think is intelligent in a highly uncertain world. I don't know how some of these young hedge fund guys do it, being 160 percent gross long and 40 percent net long. I'm not questioning anybody, but if you're running a lot of capital, to be that gross long you have to either have enormous positions where you give up liquidity or you have to have an incredible number of positions, too many to follow effectively. Our level of diversification reflects our unwillingness to make such giant bets or to give up liquidity. We could liquidate our portfolio in 48 hours.

—Leon Cooperman, Omega Advisors

With six analysts and the amount of money we have, [50 to 60 positions] has turned out to be what we feel we can best manage. It's not more concentrated out of prudence and humility. There's always a chance we'll be wrong on any given idea.

—Spencer Davidson, General American Investors

Part of our rationale [for holding more than 300 positions at a time] is just the practical reality of running $8 billion in a small-cap strategy—with that much money, you can't hold 50 stocks without moving well out of smallcap range for many of them. Another practical consideration is our investor base, which is retail investors and large institutions. Performance obviously matters when they choose T. Rowe Price to run their small-cap assets, but they also want to be comfortable that the portfolio isn't going to blow up. For many investors volatility is the enemy of rational investment decisions, so the less volatile we are, the more likely our investors won't sell at the bottom and buy at the top. Running with the level of diversification we have, the standard deviation of our returns has been lower than that of our benchmark Russell index. Philosophically, I find broad diversification makes it easier to be a contrarian. We made the mistake in 2007 of buying some housing, recreational vehicle, and mobile home stocks after they fell 50 percent, which clearly turned out to be too early. But because of the way we run the portfolio and our recognition of the risks involved, we never made those holdings, in aggregate, more than 3 percent of the portfolio. While that particular out-of-favor bet hasn't paid off, it hasn't hurt us much either. As long as the potential upside is high, we should be making those types of investments and they can make a real difference when they work.

—Preston Athey, T. Rowe Price

Small-cap stocks by definition are more fragile, more likely to have one dominant product or one key executive or one big customer. Strange things happen, so you have to diversify no matter how much you may love individual names. When something strange happens in one of Johnson & Johnson's or GE's businesses, it's a rounding error to the overall company. In a small-cap it can blow it up, so you don't want to be overly exposed in any one name.

—Whitney George, Royce & Associates

Concentration and micro caps don't mix well, so we typically own around 100 names, with a big position being 3 to 4 percent of the portfolio. Tiny companies are by definition more vulnerable to catastrophe if something goes wrong, so we try to limit the potential damage from that by owning a lot of them. I've had people ask if we're spreading ourselves too thin by owning so many positions at a time. What I answer is that there's an enormous difference in the effort required to follow a big company than a small company. I'd argue that a portfolio of 20 large-cap companies, each of which is in five or six distinct businesses, is more difficult to keep track of than 100 small companies that typically operate in a single niche. An IBM or a Disney can have a single footnote longer than a lot of the entire annual reports I look at.

—Paul Sonkin, Hummingbird Value Fund

Our level of diversification [130–140 positions] is just spreading the risk. It serves us well during downturns, which was certainly reinforced in 2008. I think it also makes us less emotionally attached to ideas and more willing to admit we're wrong, which is important for any investor.

—Tom Perkins, Perkins Investment Management

Our flagship mutual fund today has 300 stocks. Buying things when they meet the valuation characteristics that have worked for us in the past is our selection methodology, period. It's not about picking the best 5 percent, 10 percent, or 20 percent of those—I don't know which ones those are. We've done a distribution analysis of our winners and losers: 18 percent of our stocks have lost 50 percent or more, while 25 percent have made 250 percent percent or more. The math has worked in our favor by exposing ourselves to as many multibaggers as possible that meet our valuation criteria at the outset, and then patiently waiting.

—John Buckingham, Al Frank Asset Management

While we've generally avoided being hurt by underhanded executives, that risk is always there and it's far more pronounced if you're running a concentrated global portfolio. A second reason we're more diversified [with up to 150 stocks] is because I believe a lot of our alpha comes from being in the right sets of companies rather than the right specific

companies. If we get the themes right, we'll do as well, with lower volatility, owning more names rather than fewer.

—Oliver Kratz, Deutsche Asset Management

The knock on funds as diversified as ours is that they're index-huggers, which given the geographic breadth of where we invest, is not at all the case for us. I know the argument that you should only own your best 30 or 40 ideas, but I've never proven over time that I actually know in advance what those are.

—Jean-Marie Eveillard, First Eagle Funds


A corollary to the determination of how many positions generally to hold is how to size those positions relative to each other. Tolerance levels for larger position sizes obviously vary, but even the most concentrated investors at some point typically respect the admonition to not put too many eggs in one basket.

* * *

We believe in constructing the portfolio so that we put our biggest amount of money in our highest-conviction idea, and then we view the other ideas relative to that. We find things that we think are exceptional only occasionally. So if we find something that is really set up, where we think it's mispriced, where we have a good understanding of why it's mispriced, where we think the mispricing is very large and the overall risk is very small, we take an outsized position to make sure we give ourselves the chance to be well compensated for getting it right.

—David Einhorn, Greenlight Capital

We're in the camp that there just aren't that many good ideas and when we identify one, we want to make sure it can have a meaningful impact on performance. The biggest holdings are those in which we have the most conviction, which is a function of several things: the size of the discount, the potential for intrinsic-value growth, having a clear and strongly held variant view, identifying a meaningful catalyst or catalysts, liquidity, and the extent of the positive impact on portfolio diversification.

—Steve Morrow, NewSouth Capital

Perfect investments have three layers of return. The first layer is the short term return to what I'd call static intrinsic value. The second one is when the business, strategy, and management turn out to be what you think they are and there's real value creation. The third layer, if you're really lucky, is when the market gets so excited that it discounts more and more of the future into the present. The big homeruns are usually there. I always try to find at least two layers of potential, but it's also important to recognize when you put something in your portfolio whether it's a one layer or a two-layer name. You should make a two-layer name a bigger position.

—Lisa Rapuano, Matador Capital Management

We're looking for a total annual return of at least 25 percent, with position sizes adjusted for the degree of difficulty. For a given expected internal rate of return, the lower the outcome's expected volatility, the higher the position size. We create an estimated risk-adjusted Internal Rate of Return for everything and then allocate the portfolio based on that.

—Steven Tananbaum, GoldenTree Asset Management

In holding around 40 stocks at a time, we're trying to get the appropriate balance between diversification and putting most of our dollars in the names we like the best. As a practical matter, it's difficult to find 40 names that you really like. The lion's share of your excess returns will come from a few names—the trick is identifying which those will be and placing bigger bets on them. Our clients typically require we limit maximum position sizes to 4 percent, but even with that restriction it makes a big difference in results over time if your largest positions outperform.

—Paul VeZolles, WEDGE Capital

Our position sizes are set based on how well each company fits our three investment criteria [valuation, business quality, and balance sheet strength]. If it clears each hurdle with flying colors, it will be at the top end of the portfolio in terms of position size. If, say, it's cheap and the business economics are fine, but it just clears the hurdle on financial soundness, it will be at the bottom end. That gives a risk/reward profile to the entire portfolio—it's perfectly fine that our deepest-discount stocks may not be our biggest positions.

—George Sertl, Artisan Partners

Our positions tend to be equally weighted. We know there are potential errors in the portfolio, which we'd obviously avoid if we could predict what they were. Since we can't, we assume the future errors are randomly distributed, which is a primary reason we equally weight the positions.

—Bernard Horn, Polaris Capital

My view is that whatever edge I have comes more from knowing where to shop than knowing specifically which of the items I buy will be the best. So I maintain roughly equal stakes to reflect that.

—Ralph Shive, Wasatch Advisors

For each position we define a downside price at which the stock would trade if everything about our thesis turned out to be wrong. In deciding whether to put something into the portfolio, we'll assign probabilities and look at the expected value, but the downside is particularly important in sizing the position. We don't want to lose more than 100 basis points in return in any one position, so if our downside is 20 percent below the current price, say, we'd put on no more than a 5 percent position.

—Curtis Macnguyen, Ivory Capital

Our calculated downside price is extremely important in how we size positions. We limit each position to a maximum risk, measured in basis points, to our downside price. In other words, if a stock went to its downside price, we don't expect the fund to lose any more than the maximum risk we've defined for that particular position.

—Jeffrey Smith, Starboard Value

Being concentrated doesn't mean we'll just take the 15 best ideas we have and plug them into our portfolio. Because of that level of concentration, the companies we choose will overall likely be less cyclical, with more stable underlying business models. We at the margin will be less apt to hold names with higher expected values if that coincides with much larger downside risk.

—Lee Atzil, Pennant Capital

In periods of rapid change in liquidity and economic conditions, the odds that we're simply wrong about our estimates of companies' near-term fundamentals are higher than average. As a result, we're more focused then on maintaining flexibility—through cash levels and buying power—and in sizing our bets according to the medium- to lower-confidence environment we're in. We won't necessarily make fewer bets, but they'll be smaller in size.

—Larry Robbins, Glenview Capital

What tends to happen is that as the market gets more expensive we take on more, less discounted names, and when the market is less expensive, we'll have fewer names trading at bigger discounts. At the market peak in 2007, for example, we held about 40 names in our large-cap portfolio and the overall price-to-value ratio based on our estimates got to 82 percent. The other extreme was March 2009, when our weighted-average price-to-value ratio got down to 40 percent and we concentrated the portfolio in 18 names.

—C.T. Fitzpatrick, Vulcan Value Partners

Our fund usually has 40 to 60 positions. The actual number at any time is usually a function of how pricey the market is: When discounts are larger, the full weights tend to be 2 to 4 percent; when discounts aren't as large, position sizes are more like 1 to 3 percent.

—Eric Cinnamond, Intrepid Capital

I will not put more than 5 percent of the portfolio in any stock, and we usually don't have more than 2.5 percent. Early in my career I had 20 percent of my portfolio in Johnson & Johnson just before Tylenol was laced with poison. My objective is to produce an above-average, long-term return, and I think I can do that without taking that kind of concentration risk. Things happen. If I really knew the best stock in my portfolio I'd put 100 percent of the portfolio in it, but I don't. [Financial columnist] Dan Dorfman once asked me in an interview what the best and worst stocks were in my portfolio. I told him the worst stock was Converse, the shoe company, which he dutifully reported in his column. It got taken over two days later, up 50 percent.

—Robert Olstein, Olstein Capital Management

Our not letting single positions get to more than 5 percent of the portfolio is a function of having seen the stocks of too many good companies fall off a cliff as a result of something out of left field. One example I like to use is when Merck announced it was pulling Vioxx off the market in 2004. You went to bed with the stock at $45 and woke up with it 25 percent lower. I'll give up some of the upside you might get from an outsized position in order to be better protected from a risk you couldn't possibly foresee.

—Robert Kleinschmidt, Tocqueville Asset Management


A portfolio's level of concentration or diversification can clearly go beyond just the number of stocks held, also encompassing how apt holdings are to move in concert as a result of market moves or broader macroeconomic and industry trends. To some investors this is critical input into the portfolio's risk/return profile, while to others it's only a passing reference.

* * *

We have what we call a risk-bucket model. We look at the sensitivity of each of our holdings to several macroeconomic factors: Is it economically sensitive or recession resistant? Is it hurt or helped by increases in energy prices or interest rates or the dollar? Does it have political or regulatory risk? By assigning positions to any appropriate buckets, we can better understand the extent of the risks we're taking on a portfolio level. There aren't any triggers or limits, but we need to know our exposure to things like widening credit spreads or a higher dollar's impact on exports, and then be comfortable with that exposure. This comes most into practice when we're considering a new buy and want to know its potential impact on portfolio risk and diversification.

—Steve Morrow, NewSouth Capital

We tag every stock in our portfolio for more than 40 possible spread-risk factors on which stock prices can diverge dramatically. The factors include common ones like sector exposure, market cap, liquidity, leverage, and dividend rates, and maybe less obvious ones like exposure to China or the constitution of the shareholder base. At any given time, for example, we'll know that 16 percent of our longs and 13 percent of our shorts are in highly leveraged companies. We'll know our exposure to companies that should perform well in an inflationary environment versus those that won't. We focus on managing the spread risks between our longs and shorts so that we don't have significant exposure to unintended bets. We want our returns to derive from our skill as analysts and not from all the other factors that can create price volatility. In other words, the goal is for our longs and shorts to move relatively in sync with each other while we wait for fundamental catalysts to revalue our longs upward and our shorts downward.

—Curtis Macnguyen, Ivory Capital

We're keenly focused on how our holdings line up as cyclical or noncyclical. We have a lot of macro concerns, but it's difficult to translate those into an investment strategy when our investment horizon for individual stocks doesn't match up particularly well with how the macro issues may play out. Our base-case position is that the U.S. economy will sustain tepid growth through the dramatic deleveraging process the country is going through. But that may play out in a relatively organized fashion over 20 years (witness Japan), or it may cause a severe crisis in one to two years (witness what's going on in Europe). Our approach in the face of all that has been to keep the portfolio somewhat balanced between cyclical and noncyclical exposures, while being tactical in moving toward or away from either.

—Timothy Beyer, Sterling Capital Management

I never pay attention to sector or industry concentrations—I don't believe it's a reliable tool for diversification. Enron and the banks that lent to Enron were in entirely different sectors, but their fortunes were tied by that relationship. If I own Nestle, am I geographically diversified by holding a company that has its headquarters in Switzerland but earns almost none of its revenue there? I do pay attention to codependencies of outcomes between companies and think true risk reduction comes from purchasing securities that are inversely or noncorrelated. For example, owning natural gas producers, which benefit from a rise in natural gas prices, and also holding naturalgas- based utilities that benefit from a price fall would reflect hedging for inversely correlating outcomes. An example of noncorrelation is the relationship that mostly exists between the economic climate and the volume of securities traded on exchanges. While the economic climate may impact many securities we hold, the success of the publicly traded exchanges we own is largely independent of it.

—Murray Stahl, Horizon Asset Management

We pay attention to end-market diversification, within our companies and across the portfolio. One large holding in a company with five separate global financial businesses is probably more diversified than five holdings in similar regional bank stocks. Our goal is to own businesses with uncorrelated enough end markets that we can continue growing the intrinsic value of the portfolio in any kind of market.

—Brian Bares, Bares Capital

Long/short funds typically don't blow up because they made a bunch of wrong fundamental stock picks. They blow up because they're overexposed to correlated sectors, or they own too many leveraged companies, or they have too many illiquid positions. These are explanations you see all the time in funds' letters to investors. That's exactly what we try to avoid.

——Curtis Macnguyen, Ivory Capital

Our rule is to own something in every sector, in part to avoid missing something important because it's out-of-sight, out-of- mind. We're not a slave to our benchmark—the Russell 2000 Value Index—but I typically don't go much below half, or much above twice, the index weighting in any sector. I've found that gives us plenty of room to beat the index, while avoiding the type of relative volatility that makes most investors nervous.

—Preston Athey, T. Rowe Price

We have sector limits, at plus or minus 10 percentage points from the percentage weighting of the 10 S&P 500 industry sectors. That gives us the flexibility to zero out sectors that are less than 10 percent of the index— like utilities and telecom today—or overweight fairly heavily in sectors we find attractive, such as energy. But it does put limits on how under- or overweighted we can be, which we think provides prudent diversification and risk management.

—Daniel Bubis, Tetrem Capital

One of our biggest mistakes was ten years ago, going too heavily into emerging market closed-end funds, which were selling at 25 to 30 percent discounts. When the Russian debt crisis hit, the NAVs got hammered. It's one of the first lessons you learn: be diversified enough that if that 1-in- 100 event happens, you don't blow up.

—Phillip Goldstein, Bulldog Investors

In our flagship domestic and international products we do not hold individual positions over 5 percent and will not have more than 30 percent in any one sector. Historically, whether it was energy in the 1980s, technology in the late 1990s, or financials more recently, when a single sector approaches 30 percent of our portfolio or of the market that signals the end rather than the beginning of great investment performance.

—Jerry Senser, Institutional Capital LLC

We cap a given industry's exposure at 25 percent of the portfolio, which is a check on the innate lack of humility we often have as investment managers. Owning five or six 4 percent positions in an industry is a good, strong bet, but also isn't betting the house on how smart we are relative to everyone else.

—Jeffrey Bronchick, Reed, Conner & Birdwell

People tend to assume that the only form of active portfolio management is through relatively concentrated portfolios. We think there's an equally legitimate form of active money management in running a diversified portfolio that has nothing to do with the benchmark.

—Charles de Vaulx, International Value Advisers

We don't benchmark at all. I don't care if we own almost no financials and I don't care if we own an excess amount of energy. We'll go where we think the value is and let the weightings fall where they may.

—Steven Romick, First Pacific Advisors

We've purposely avoided basing our bonuses on performance against benchmarks. We're always running into managers who say they're unable to look at certain stocks because they don't fall within a prescribed benchmark. They tell us, “I can't take the risk. If I buy it and it goes down, I'll have to write all sorts of memos explaining it, and I'll get less bonus because my portfolio went down more than the benchmark.”

—Bernard Horn, Polaris Capital

You can understand why many succumb to the pressure to hug the index, so to speak. But we believe if you go down the road of trying to make sure you'll never do much worse than the index, you're almost insuring that you'll never do well enough to justify your compensation as an active manager.

—Bill Nygren, Harris Associates

The only way to add value as an active manager is to be persistently different than the index. We tell prospective clients that if their main goal is to minimize standard deviation around the index, save money and buy an index fund.

—Jeffrey Bronchick, Reed, Conner & Birdwell

No comments:

Post a Comment