Case StudiesSuperinvestorsThink Differently

Portfolio Turnover–A Vastly Misunderstood Concept

A while back I wrote a post about how the gap between 52 week high and low prices presents an opportunity for investors in public markets.

I mentioned that this simple observation (the huge gap between yearly highs and lows) is all the evidence you need to debunk the theory that markets are efficiently priced all the time. I think the market generally does a good job at valuing companies within a range of reasonableness, but there is absolutely no way that the intrinsic values of these multibillion dollar organizations fluctuate by 50%, 80%, 120%, 150% or more during the span of just 52 weeks.

The market is constantly serving up opportunities. I just checked a screener and there are 375 stocks in the US that are 50% higher than they were 1 year ago today.

This leads me to a thought that I think, for some reason, is not really discussed in investing circles—at least not in value investing circles: and that is the concept of portfolio “turnover”.

To think about portfolio turnover, let’s first take a look at a concept that security analysts and value investors think about more often: asset turnover.

Asset turnover basically measures how efficient a company is at using the resources it has to generate revenue. It’s simply a company’s revenue in a given period divided by its assets. Generally speaking, asset turnover is a good thing—the higher the better. If two companies have the same asset base, the company with the higher level of sales is doing a better job at employing those assets.

Coke and Pepsi

Coke and Pepsi are somewhat similar businesses, but it isn’t necessary to compare their business models when it comes to understanding the math of turnover. Just look at how Coke’s profit margins are almost double the margins at Pepsi, but Pepsi is about equally profitable (produces similar returns on assets) because Pepsi is more efficient than Coke is at using the assets it has.

Let’s glance at two homebuilders:

NVR and Lennar

NVR has a different business model than Lennar as it uses less capital (it employs a smaller asset base). This allows NVR to be almost three times more efficient with its resources than Lennar, and although Lennar has a higher profit margin, NVR produced a much better return on assets.

We can compare two businesses in different industries to see how their business models and operating results affect their profitability:

Coke and Whole Foods

Coke and Whole Foods produce roughly the same ROA, but got there in very different ways… Coke has very high profit margins but takes nearly 2 years to produce $1 of revenue for every $1 of assets. Meanwhile, Whole Foods’ profit margin is less than 1/4th the size of Coke’s, but it turns over its asset base nearly 5 times faster, yielding roughly the same return on the resources it has to deploy.

These examples aren’t to say one business or one measurement is better than the other–it’s really just to point out the importance of turnover.

Inventory turnover is a similar ratio. A grocery store is a very low margin business, but in some cases grocers can produce adequate (or sometimes better than adequate) returns on capital if they are able to turn their inventory (merchandise on the shelves) faster than competitors. Two competitors with identically low profit margins might have vastly different profitability because one grocer might be producing much higher ROA due to its ability to turn its inventory over faster.

So in business, it is clear that asset turnover (and inventory turnover) is a good thing. The higher the turnover, the higher the returns.

Portfolio Turnover

I once mentioned I have put together notes on investors who have achieved exceptional (20-30% annual returns or better) over a long period of time (say 10-15 years minimum). There are a variety of strategies and tactics employed, but there are a few common denominators. In addition to the expected commonalities (most are value investors), there is one common denominator that isn’t talked about much: portfolio turnover.

Portfolio turnover is a phrase that I’m using—I don’t like using phrases and words that you might find in a CFA textbook, but this is the easiest way to refer to the concept. Basically, think of portfolio turnover as asset turnover.

The capital you have in your account might consists of stocks, bonds, cash, etc… these are your assets. The faster you turn these assets over (at any given level of profit), the better.

It’s simple math. I think a lot of value investors get hung up on the Buffett 3.0 version. Let Seth Klarman explain this… Klarman once said that Buffett’s career has evolved a few different times and can be categorized generally as follows:

  • Stage 1: Classic Graham and Dodd deep value and arbitrage (special situations)
  • Stage 2: Great businesses at really cheap prices (think American Express after Salad Oil Scandal, Washington Post, Disney—the first time at 10 times earnings in the 60’s)
  • Stage 3: Great businesses at so-so prices

Now, if we look at Buffett’s results, even lately, some might take issue with Klarman calling them “so-so” prices. But nevertheless, I think Klarman is basically correct in his assessment of Buffett’s career, and I actually think Buffett himself would agree with this. As Buffett’s capital base expanded, he had to begin to begrudgingly adjust the investment hurdle rate that he required. He mentions this in his 1992 letter to shareholders, replacing his demand for “a very attractive price” with simply “an attractive price”.

This description by Klarman took place during an interview with Charlie Rose, and Klarman jokes that he (Klarman) is still in Stage 1, scavenging for bargains. What’s interesting about this comment, is Klarman has been able to produce really solid returns on a very large amount of capital, and I think it’s in large part because of the simple math of asset turnover—Klarman buys bargains, waits for them to be valued at a more reasonable level, sells them, and repeats.

Walter Schloss was another master at turning over his portfolio that was filled with bargains. Schloss actually ran his portfolio like a grocery store. I’d say on balance, his stocks produced relative small profits (I’d venture Schloss had many 20-50% gainers, but very few 5-10 baggers), but collectively, he produced 20% annual returns for nearly 50 years because he was able to adequately turn over his “inventory” (i.e. his stocks) fast enough. This isn’t to say that you have to look for activity, or actively trade—Schloss said he kept his stocks an average of  3-4 years. But it just means that he would not have produced anywhere near the results he did if he held his stocks “forever”, or for 10 years instead of 4, etc…

Walter Schloss was akin to the low margin grocery store that didn’t produce exciting margins on any one product, but collectively across the store it was able to effectively turn over the merchandise fast enough to make exceptional returns on the assets it employed.

Some other investors might be more akin to the higher margin, lower turnover businesses that might produce much lower asset and inventory turns, but still generates very attractive returns on capital because of its very high return on sales (profit margin). These investors hold stocks for longer periods of time, but find big winners that rise 3, 5, 10 times in value over many years. A business can produce incredible profitability on lower asset turnover if it can wring out a large amount of bottom line earning power from its top line revenue.

The Two Drivers of Profitability

Some people might be familiar with the simple math of this situation, but it might help to briefly illustrate this to show what ROA (Return on Assets) consists of:

The Return on Assets (ROA) is one measure of profitability and it is calculated simply by dividing net income into total assets. A lemonade stand that produces $20 of earnings and has $100 worth of assets (the stand, the small square of front lawn, the inventory of lemonade, etc…) is producing a 20% return on assets.

The two functions that determine ROA are:

  • Profit Margin (some accountants refer to this as “Return on Sales”). Profit margin is calculated as follows:
    • Profit Margin = Net Income/Sales
  • Asset Turnover (the measure of how efficiently a business uses its assets—i.e. how much revenue can be generated from each $1 of assets). Asset turnover is calculated as follows:
    • Asset Turnover = Sales/Assets

So I hear a lot of people talking about the profit margins (big winners) but few investors talk much about asset turnover (how quickly you go from one investment to the next). And worse yet, when they do discuss turnover, it’s usually negative (most investors say lower turnover is better, which is not true—more on this shortly).

Higher Turnover Isn’t Necessary—But It Does Influence Returns

Now, it’s important to keep in mind the above equation—there are two drivers of profitability of a business:

  • Efficiency—how much revenue you can produce from your available resources (assets)
  • Profitability—how much profit you earn for each $1 of sales

So, turnover (whether we’re talking about asset turnover in the context of a business, or portfolio turnover in the context of an investment account) is just one driver of the returns that the business (or portfolio) generates.

The other driver is how much money you make on each $1 of sale (or each $1 invested).

If your business begins to turn over its assets more slowly (i.e. it begins to generate less revenue per $1 of assets), then you’ll need to make up for that by earning a higher profit margin on each $1 of revenue if you are to maintain the same ROA.

Similarly, as an investor, if your portfolio turnover decreases (which is often the result of a longer time horizon), your profit margin (in the context of investing, the amount of money you make on each $1 invested) must increase if you are to maintain the same level of annual returns on your overall portfolio.

I think this is where there is somewhat of a disconnect in the value investing community—which often considers portfolio turnover to be a negative thing. In and of itself, turnover is not bad. In fact, generally speaking, the math tells us that it is one of two main drivers of investment performance. So it’s actually necessary!

Why Do Investors Think Portfolio Turnover is Bad?

I think the reason for this negative connotation is that portfolio turnover is often associated with excess, or inappropriately high levels of trading, which is often done for emotional reasons without regard for the fundamentals of the business.

But let’s assume you are a rational, disciplined value investor. If that’s you, then you should try to turn your portfolio (your assets) over as fast (as efficiently) as possible. The faster you can buy and sell 50 cent dollars, the higher your returns.

Again, simple math (this might be painfully obvious, but I’m still going to demonstrate):

Let’s say you buy a stock at $10 and you sell it at $20:

  • If it takes 5 years to get from $10 to $20, you’ll earn a 15% CAGR on that invested capital
  • If it takes 2 years to get from $10 to $20, you’ll earn a 41% CAGR on that invested capital
  • If it takes 1 year to get from $10 to $20, you’ll earn 100% CAGR on that invested capital

In this case, your “profit margin” is the same in each case: it’s 100% in all three examples (the stock doubled in all three cases). However, your CAGR increases as your asset turnover increases—in other words, the more opportunities like this you can find and the faster they play out, the higher your portfolio returns will be.

So that demonstrates the various CAGR’s on the same level of profit margin. This is the same basic math that you’d see if you compare two companies with a 10% net margin, but Company A turns over its assets twice as fast as company B, then Company A’s ROA will be twice as high.

Now let’s quickly look at the same level of asset turnover on different levels of profitability:

Let’s say each year on January 1st, you buy one stock, and each year on December 31st, you sell it to buy something else:

  • If the stock you bought goes up 15% over the course of the 1 year, obviously your CAGR is 15% on this 1 year investment
  • If your stock goes up 25%, your CAGR is 25%, etc…

In this example, your asset turnover is exactly the same (you turn over your assets once per year in this case), but your profit margins are different.

Obvious stuff, right?

I think so, but I consistently read a lot of people referring to portfolio turnover as a bad thing, which runs counter to the math behind these examples.

Buffett’s Returns and Peter Lynch’s Famous 10-Baggers

It’s clear to see with these simple examples that portfolio returns (ROA) is dependent on two drivers:

  • How much you make on each investment (profit margin)
  • How quickly you can turn over your assets (asset turnover)

Buffett’s transition that Klarman referred to above is one where he transitioned over the course of his career from a lower profit margin, higher turnover investor to a higher margin, lower turnover investor. In the 50’s and 60’s, Buffett made many more investments, and made much smaller profits on average (in other words, he bought stocks, sold them when they appreciated to buy still more undervalued stocks). In the 80’s and 90’s, he began making fewer investments (due to increasing capital levels), but his profit margins grew (he went from making 20%-50% gains in shorter time periods to making 1000%+ gains over many years).

Interestingly, Buffett’s results (on a percentage return basis) were much better when he had higher turnover (and lower average profit per investment) in the early years than they are now. In fact, Buffett said his best decade of returns was the 1950’s, when he was making 50% annual returns, and investing in a variety of bargains and special situation events.

This wasn’t necessarily intended or by design, it was simply that Buffett was “taking what the defense gave him”. As his capital grew, he had to look for larger investments and had to extend his time horizon.

If he were investing again with $1 million or so, he’d be making many more investments and his asset turnover would be much, much higher—there is absolutely no doubt about this.

He may have a few investments that become big winners, but there would be very few 10 baggers, and many, many smaller, faster gains.

One other point that runs counterintuitive to what most people think: Peter Lynch is famous for the term “10-baggers”—investment that rise 10 times in value. But in fact, when Lynch started running the Magellan fund and was producing incredible 50%+ returns in the early years, his turnover exceeded 300% every year for the first 4 years (in other words, the average length of time he held a stock was only 4 months). I think in reality, his fantastic track record is much more because of higher portfolio turnover and much less because of the famous “10-baggers” that he cites in One Up on Wall Street.

Certainly profit margins are just as important a driver to profitability (portfolio returns), but I think turnover is vastly misunderstood.

I think it’s important when listening to the great investors–even Buffett–to keep in mind this math when you hear ubiquitous investment advice and generally accepted wisdom regarding turnover, investment time frames and holding periods.

52 thoughts on “Portfolio Turnover–A Vastly Misunderstood Concept

  1. John, outstanding article. I have thought about this very thing many times. I started out doing the same thing, buying cheap stocks with higher turnover and did very well and enjoyed it. Then, after studying Buffett and many others who were Buffett followers decided to concentrate my portfolio and really “own the business”. I studied conference calls, in-depth research on companies, trying to decide which companies were great to own long-term, etc. I ended up driving myself nuts and I have gone back to more investments with higher turnover and having a lot more fun. I of course would love to find the 10 baggers but they are so hard to find, whereas good bargains are pretty routine. Plus you give yourself room for a few lousy purchases. Thanks again and keep up the good work!

    1. Thanks for the comment Jon. I think it’s all about each individual situation. Some businesses can continue to generate high returns for a period of years, but often times an undervalued stock reaches fair value sooner than later, and then subsequent returns are mediocre. Each situation is different, but I think it’s important to keep the concept of turnover in mind when looking at ideas. I think all the same principles that Buffett talks about are relevant, but I think if you’re running a smaller (i.e. less than billions) portfolio, it probably will behoove you to think more like the 50’s and 60’s Buffett vs. the 90’s and 2000’s Buffett.

      All of his advice is good, it just has to be put into context.

      1. Hi John, I thought it was interesting that Buffett’s latest letter said that buying “cigar butt” companies was an excellent short term strategy but wasn’t scalable. For those of us with much smaller sums of capital, “cigar butts” might be the way to go (with the occasional compounder when they are available at good prices). Interesting to think about in light of your article!

        1. Yes Jon, that’s a good point. I think it’s important to keep an open mind to any and all investment ideas that you a) understand, and b) can reasonably estimate value. Some of these might be bargains, others might be great businesses, etc… Buffett always liked compounders, but he’s also been fond of more quantitative bargains throughout his career also. He bought baskets of Korean stocks in 2005 in his personal account that were extremely cheap (3-5 times earnings)–an experience that said brought him back to his youth. I think the key is just keeping things simple and focusing on ways to invest capital where you can easily see a gap between price and value. Restricting investment decisions to those investments that only meet that simple filter will go a long way I think.

  2. I’m not sure you give enough attention to how this analysis differs based on whether an account is taxable or not. High turnover strategies are definitely more fun for non-taxable accounts.

      1. Yes you had a thoughtful reply re taxes. My personal observations:
        – It took me a long time to appreciate the impact of taxes. I’m 40 and have been investing professionally for around 15 years. Its only in the past few years that I look back and realize the compounded impact of all those taxes. Its a little like transaction costs/mgmt fees/etc… – in any one period they seem de minims but its that compounding impact over time which really adds up.
        – Yes if you have a strategy returning 30-50% then you do it regardless of taxes. But I think of this topic the same way as I think of value vs growth investing. If one had a crystal ball and could identify long-term hyper growth companies ex ante then you would say to hell with value and pay almost any multiple for them today. But in the real world where thats really really hard to do then value investing is the best practical strategy. If I knew a high turnover strategy would get me early Buffet returns I’d just do it. But since I think achieving those returns is pretty darn hard&unlikely I try to control things like taxes where I can.
        – Let me conclude with a admission that I am far from single minded on this topic. When markets keep surging and you are paying taxes along the way taking gains you feel like a sucker. But I have too often let a position linger around after its hit a full valuation to avoid taking taxable gains and truly regretted it later when something happens and I have no margin of safety. I don’t know where the perfect balance is if there even is such a thing.

  3. I’m a big fan of the blog John. I enjoyed reading this post but wish you had addressed a couple points . The first and most obvious point is the tax implications of high portfolio turnover. The next and equally important point is the difficultly of behaving rationally when your goal is to turn the portfolio over more frequently. It is easy to lower the bar for new investments and buy businesses that are of a lower quality than the long-term investor. I think this is especially concerning at this point in the cycle. Keep up the great work! Thanks

    1. Hi Matt, thanks for the comment, and thanks for reading.

      I think I need to correct a few things that may be getting misinterpreted. It shouldn’t be your “goal” to achieve a certain level of turnover (I realize I imply this when I say higher turnover is better at a given level of profit margin, which is true).

      But your goal as an investor should simply be to locate bargains and patiently let them reach intrinsic value. You can’t force things, and as you say, you have to be very aware of behavioral flaws and you have to be very disciplined.

      The post was really just pointing out the math of the situation, and to say that turnover is not bad. You shouldn’t aim to achieve a certain level of turnover, just aim to find as many bargains as you can.

      The more you find, the more turnover (and the higher your returns) will be.

      As I mentioned in the post, this was Buffett in the 1950’s. He had the highest returns of his life that decade and also the highest level of turnover–simply because he was finding many bargains. As it became harder for him to find as many bargains (because of higher levels of capital mostly), his holding period expanded, his turnover dropped and so did his returns (although it must be nice to have such problems).

      Regarding taxes, in my opinion people think about taxes too much. More emphasis should be placed on trying to find investments and making money, and let the taxes take care of themselves. At a certain turnover level, the tax benefits mostly disappear anyhow. See below for more comments on this.

  4. Great article John.

    It is a combination of banking on mr market’s generousity and the underlying long term performance of your companies. Luckily there’s money in both. I would think though that it takes more skill (and a wide range of knowledge) to be successful with more turnover. Also, if you get it right and buy an otusatanding company, then you don’t have to do anything. That is my ultimate goal. And there’s tax advantages in that, per Munger:

    “Another very simple effect I very seldom see discussed either by investment managers or anybody else is the effect of taxes. If you’re going to buy something which compounds for 30 years at 15% per annum and you pay one 35% tax at the very end, the way that works out is that after taxes, you keep 13.3% per annum. In contrast, if you bought the same investment, but had to pay taxes every year of 35% out of the 15% that you earned, then your return would be 15% minus 35% of 15% or only 9.75% per year compounded. So the difference there is over 3.5%. And what 3.5% does to the numbers over long holding periods like 30 years is truly eye-opening. If you sit back for long, long stretches in great companies, you can get a huge edge from nothing but the way that income taxes work.
    This isn’t exciting to talk about, but it is where the rubber really meets the road in terms of investment success. 3.55% per year outperformance after taxes is incredible, and you don’t have to pay a hedge fund “2 and 20? to get it.”

    1. Thanks Daniel. I just noticed this comment, and thanks for sharing the Munger quote. I might write a post on this comment, because what Munger is talking about makes sense, but it involves a 30 year time horizon. The tax implications (benefits) go down significantly if you did this investment every 10 years, and even more every 5 years, etc…

      So I think this is another misunderstanding when it comes to taxes–the tax benefits on your overall portfolio go down drastically once your turnover reaches 30%, 40%, 50% or thereabouts. For your entire portfolio to achieve the tax benefits that Munger is talking about, your portfolio would have to have 3% turnover (you’d have to hold every stock for 30 years without selling and they’d all have to do 15% CAGR). That’s unlikely. And as your portfolio turns over faster, the tax benefits decrease dramatically. This isn’t a problem though because unless you happen to put your entire portfolio into the 1 or 2 stocks that can compound at 15% for the next three decades, you’re better off buying bargains, selling them when they reach fair value, paying the tax, and repeating. It will certainly be difficult to beat 13.3% after tax, but the other alternative is to buy 1 or 2 stocks that you think will do 15% annually for 30 years. I wouldn’t want to have to do that, as I think it would be virtually impossible to be confident enough to put all your money into two business and expect them to do 15% per year for 30 years. And you could own 10 stocks to diversify, but then you virtually guarantee that your portfolio won’t do 15% (you won’t find 10 that will all compound at the level Washington Post did for 30 years).

      Anyhow, I’m rambling here, but these are some thoughts.

      Keep in mind, Munger was doing 30% per year for a decade plus in his partnership, and even including two -30%+ years he averaged about 20% annually. He was buying and selling many stocks, doing arbitrage deals, special situations, etc… he put 100% of his money into one special situation at one point. So I think his math is accurate, but I think it needs to be put into context, and you need to think about it in a practical way–is it possible to replicate? That sort of thing…

      I think it’s more beneficial to really study what those guys were doing when they were a) much smaller and b) producing huge annual returns.

      Their advice that they provide now (in my opinion) runs counter (in some cases) to what they actually did back then.

      1. John, thanks for this excellent blog. This has been an extremely interesting thread to read.

        I’m curious on where you got the information on Munger’s investments because I don’t remember any specifics like the ones you mentioned in Damn Right! or Poor Charlie’s Almanac.

        Thanks in advance.

        1. Hi Dan,

          I’ve pieced together Munger’s numbers from various sources, but there are a few places where his results are discussed. A few of the Buffett books have his partnership results, and Snowball has some great information on Munger that I hadn’t heard before. Schroeder talks about Munger’s early days in a few spots in that book.

  5. What about tax? I do believe one of the reasons Buffett and Munger extol a “buy and sit on your a**” investment philosophy is because of the benefit of deferring capital gains tax as long as possible. Although, to play devil’s advocate to my own question, if you have limited funds like I do, perhaps it is wisest to redeploy capital as soon as you uncover what you deem to be more promising opportunities. With the float Buffett has to invest, he has the opposite problem in that he often has far more investable funds than opportunities that are within his circle of competence. Also, I’d imagine it would be a monstrous headache trying to move that kind of money in and out of positions without moving the stock dramatically. Better to just invest in a great business and not have to worry about it. So… that was a very convoluted way of asking whether you think taxes should be considered when deciding the optimal amount of portfolio turnover?

    LOVE your articles by the way. I’ve learned a lot in a short span of time. Please keep writing! It’s appreciated!

    1. Thanks Antoine. Yeah it’s a good question. As I mentioned in the comment above, I think it pays to study the way these guys (Buffett, Munger and others) invested when they had small amounts.

      One thing I guarantee, they wouldn’t be buying Coke and sitting on it for 20 years if they had $1 million, $10 million, even $100 million to invest. They’d be buying bargains, selling them, and rarely if ever thinking about taxes.

      If you make 20-30% per year (which those guys did in their partnerships), you’ll happily pay the tax.

      The reason they hold longer now is exactly as you say–they can’t find many ideas, and when they do, they have to “buy in bulk”. It takes months to build his stake in IBM, which is one of the most liquid stocks in the index.

      As for “optimal level” of turnover–I wouldn’t think about it that way. Just let the turnover be whatever it is. Just try to find bargains and let them take care of themselves. The more you find, the more you’ll have turnover.

  6. Great article John!

    First time I came across a good explanation of ROA was Pat Dorsey’s 5 Rules for Successful Stock Investing. He goes on to add the concept of financial leverage to get to ROE. Really insightful.

    To me, the truly important thing is to understand the nature of what your’e doing. Wether it’s a low margin-high turnover, or high margin-low turnover business, you have to know what you’re playing at to have a chance of success. This applies to your style of investing to.

    Incidentally, in my view this was the genius of Peter Lynch. He was able to label (categorize) each type of investment, and therefore analyzed it using the right conceptual model (turnarounds, slow growers, asset plays…)

    The way I see it, you have to make a tradeoff. Do you preferr more but easier work (cigarbutt investing), or less but harder work (quality growth investing). Not an easy choice. When you have a lot of capital you are more or lessed forced into the second.

  7. John,Thank you for your amazing article.I read about an investor named Ahmet Okumus in Jack Schwager’s “Stock market wizards”.He is a deep value investor and have a high turnover but he made above 100% return from 1992-1998.I am from India and i can say deep value investing with high turnover work here as well.And I have a request that you should write something about position sizing,like how much percentage of the capital(maximum-minimum)to put in a single company and how much to hold as cash like stuff . Thanks again and keep up the good work

    1. Thanks Prabeeth. Yes, I’ve read that chapter on Okumus as well, and he certainly had a lot of turnover in order to compound at 100%.

      As for position sizing, I tend to be concentrated. If you were asking about Okumus, I think he put 10-20% of his money into each main idea. (He also sold a lot of options, which I think might have added some incremental value, but most of his returns I’m sure came from his main positions).

  8. John,

    Great article and a lot of interesting thoughts on ROA and asset turnover.

    One thing that would be interesting to get your thoughts on, which I don’t think you address in the post, is the real return of an account with higher portfolio turnover.

    Assuming we are talking about a taxable account, the one that trades more frequently will be paying more transaction fees and also cap gains taxes, which will lower their real return vs. an account with less activity, even one that has lower annual return. Curious to get your thoughts on those points vs. your post.

  9. Thanks for the post, I enjoyed it. A thought on something I imagine is similarly correlated with AUM and Turnover: size of investment team relative to AUM.

    One of the only reasons Klarman is able to do what he does on a large capital base is because he has built on the finest investment organization in existence so that he doesn’t source the investment ideas anymore. It’s an acknowledgement that the due diligence necessary to consistently deploy capital in that strategy is not capable of being scaled past a couple hundred million (if that). Buffett does what he does by himself, with Charlie Munger, or maybe with Todd Combs and Ted Weschler depending on how you look at it. His investment organization is minuscule compared to Klarman.

  10. I agree with you, especially because it is hard to find stocks that are going to be 5 and 10x baggers, whereas I can find many stocks that can return 50-100% in 1 to 3 years. The problem with the high turnover strategy ends up being the extra taxes that eat away at your return.

  11. John, enjoyed reading your post very much and it strongly resonates with me. Wanted to mention few points:

    (1) you are ignoring taxes as a drag on value creation and benefits of compounding (obviously, if a manager runs a portfolio comprised for non-taxable investors, the point is irrelevant). For a taxable investor the returns in “high-turnover” portfolio must be higher just to match returns on the “low-turnover” portfolio on the after-tax basis.

    (2) Nairan pointed this out already – ROA ignores differences in leverage and cap structure among companies (so adding back tax-adjusted interest expense to net income and dividing it by average assets would be a more conceptually correct way to do, in my opinion). But this is just a neat-picking since the conclusions would not change much. Plus I know from your other posts that you put the highest emphasis on ROIC vs. ROA and I bet you used ROA here for sake of speed.
    Great post, John.

    1. Thanks Artem. See below for a longer reply regarding taxes. And yes, the ROA/asset turnover analogy is really just for general comparison. The concept here is what I’m going after… I realize that it’s not a completely accurate comparison, but I think the concept is comparable.

      Thanks for reading!

  12. Thanks for the comments everyone. I noticed a number of comments/questions on taxes. That is a topic that I considered discussing in the post, but since the post was already getting long I decided to leave it for another time. But this is a topic that I also think gets overrated at times: turnover and taxes.

    To start, these things can easily be calculated, and it all comes down to the math. It is simple to compare after tax returns on two different accounts.

    Maybe I’ll put together another post with some detailed examples on this topic, but for now, I’ll say that I’ve spent a lot of time crunching numbers and thinking about the various impacts that turnover has on taxes. My conclusion is that the tax benefits that you get from lower turnover dramatically disappear once turnover gets over 25% or so. This is not exact and tough to estimate because it all depends on the numbers involved.

    But generally speaking, taxes (planning for taxes) is overrated. Any investment manager that has achieved 20%-30%+ returns has paid a lot of taxes. Very few businesses will compound their intrinsic value at 20% over long periods of time (or even 15%). The very rare ones allow you to compound your equity and defer taxes (but not eliminate them). But in order for your portfolio to achieve this type of deferral at that rate of return would require you putting all your money into the one or two stocks that happen to be able to compound at that rate over long periods.

    It’s an unlikely prospect. To achieve those types of returns, you’ll need to pay taxes, because you’ll need to sell fairly valued stocks in order to buy cheaper stocks occasionally.

    Note: I’m not suggesting you need to be trading frequently. But I am saying that if you find a stock for 50% of its value, the faster it gets to intrinsic value, the better.

    If you delay selling it for taxes, you end up costing yourself significant opportunity costs.

    The other thing to note: In the first part of the post, I commented about investors who have achieved big returns. If you’re looking for 10-15% returns, I think it’s doable to run a low turnover, tax efficient investment strategy. But your after tax returns will not come close to the after tax returns that some of the great investors achieved, because you won’t be able to invest all of your money in stocks that compound at 20%-25% annually. Even if you happen to find and invest in one or two, you’ll likely sell it at some point, and the rest of your portfolio will be compounding at a much lower rate.

    So these are just some random thoughts on this topic. We could discuss it further, but the comparison I’m using (Klarman’s 20%+, Buffett’s 30%+, Pabrai’s 25%, Lynch’s 29%, Greenblatt’s 40%), etc… these types of returns cannot be achieved without some level of turnover.

    Think about Buffett in the 1950’s. He was making 50% per year on his capital. Do you think he was worried about taxes? It was the furthest thing on his mind.

    Greenblatt made 50% per year from 1985-1994. I can almost guarantee that he never once thought about the tax implications of running a special situation fund that had higher turnover during that run (maybe he considered it if he happened to be getting ready to sell something right around the 1 year time period, but other than real minor things like that, he wasn’t concerned with taxes).

    Buffett and Munger talk about the tax benefits of owning Washington Post and making 15% compounded over many years (which works out to around 13% net of taxes if they sold it). Munger once said that this compares to 9-10% if you had to buy and sell once per year at 15% pretax. But again, this is one example and unless your whole portfolio consists of Post, then your returns will not be 15%. And even if it was all in Post, you wouldn’t compound as fast as Lynch (who had 300% turnover when Magellan was doing 40% annually in the early years) or Buffett, or Greenblatt, or Schloss, etc…

    Plus, the Munger example with Post that he discussed once involved a very long holding period. The math here is obviously correct, but to extrapolate this to your whole portfolio, you’d have to have somewhere around 5% turnover (you buy and hold each stock for 20 years).

    Again, the tax benefits dramatically fall off a cliff once your portfolio turnover climbs into the 30-40% range, which is still fairly low. And I think it’s a catch 22 because if you try to keep your portfolio turnover in the 20% or less range to really get tax benefits, then you will not compound at exceptional rates of return because the likelihood of you filling your entire portfolio with 20% compounders every 5 years is virtually impossible.

    Not to say it’s wrong to try to minimize taxes when you can. But I think making business or investment decisions that are based on taxes is usually not a productive exercise.

    Just some thoughts off the top… thanks for the interesting and thought provoking comments.

  13. John, wholeheartedly agree with your view on taxes. This is my favorite: “But I think making business or investment decisions that are based on taxes is usually not a productive exercise.” Artem

  14. Just a follow up on a comment I had in the post on Peter Lynch: Here are his first four annual returns (1978-1981 I believe) in Magellan Fund when he took over managing it:

    1978: +20%
    1979: +51%
    1980: +70%
    1981: +94%

    The turnover in each of those four years exceeded 300%.

    Now, obviously his investors paid a lot of taxes with that level of turnover. But I doubt they’d trade their tax bill for a tax-efficient 10% or 12%, etc…

    So taxes of course impact your net returns, but I think too much emphasis is focused on minimizing taxes and not enough emphasis is placed on actually trying to make a lot of money.

  15. Generally agree. As long as one compares all investment ideas by expected after-tax IRR, then it is an apples-to-apples comparison, whether the expected holding period is 10 years, or 2 months. I think the tax drag on returns also has political implications, and is why we don’t see so many fiscally liberal investors.

    One point that could have made this post better would be mentioning how the turnover and margins are terms from the DuPont model.

    Another point is that the higher the turnover rate, the more likely one is to experience adverse selection (i.e. “pulling up the grass and watering the weeds”). The reason is that if one is selling companies once they hit return rate targets, even if you choose really high targets, the remaining part of the portfolio will more likely be the companies that are suffering more problems than were expected. The solution for that I think is just to be aware of it and occasionally clear out investments that are experiencing worse-than-expected business results.

    A final point is that if one is selecting higher turnover on a particular portfolio position, then one has to be highly dedicated to not fall victim to behavioral, and emotional biases, and short-term thinking. Personally I find it easier to keep track of many different statistics about positions, and use those to inform whether I will keep a position or exit it, than to just make subjective or “gut” decisions.

  16. Hey John!
    Great article. I wanted to give my 2 cents on the tax issue. Now I see that the readers and you have dealt with it already.
    Good work

  17. John,

    I hate to say this but I really enjoyed reading this article. The reason why it is difficult for me to admit it is because I’m fundamentally a low-turn over type of investor, seeking higher margins over time. However, I acknowledge your point about Lynch and Buffet’s early-career returns, I have read a lot of books on them and what you said makes a lot of sense in my mind.

    That said, one of the most significant risk I encounter when investing is something I would name the ”picking” risk. Indeed, since my strategy implies long concentrated bets, I cannot afford to hit a bad pick with poor management, wrong strategy etc.

    I usually don’t like to take on smaller positions because to me it typically represents less conviction in my position (I do a lot of Due Diligence before acquiring shares of a company). Also, increasing the number of picks would increase the chance of hitting a bad one once again. This could be debated with the virtue of diversification included in the higher turnover strategy (implying higher turnover of smaller positions which would limit the aggregate downside of wrong picks).

    All in all, I do not want to oppose to what you have put in light in your article, just adding my 2 cent. You’re thoughts have forced me to rethink about my strategy and the way I approach the composition of my portfolio and I wanted to thank you for that.


    1. Thanks Seb. Yeah I think that the challenge of course is finding ideas. And certainly there is nothing wrong with finding the compounders, as long as you are doing a good job at identifying them. As for the number of ideas, I actually think you don’t need that many. Even if you turn over the portfolio every couple years, if you run a concentrated portfolio you shouldn’t need more than a few ideas each year. I think Buffett put together such an incredible track record in his partnership because he was brimming with ideas. He had higher turnover because he had a lot more ideas than he does now (because of capital size). I think Greenblatt did well because he ran a concentrated portfolio and didn’t need so many ideas. So there are different ways to accomplish the objective of outstanding results.

  18. interesting, I think about the topic in the same way…you missed one pretty important point…investors aren’t just investing money, they invest time too. If you are turning over a lot, you are shooting yourself in the foot because the amazing thing about investing is you can make one decision that takes a week of research and get paid off for decades. A local fund manager I know made one call in the early 1990s, buy Fastenal, their funds outperformed largely due to this one call and they now manage $40bn…a lot of other stuff at play of course but remarkable all the same. I know high turnover people who have done well too, it is just requires a completely different skill set…usually expertise in a certain region/industry. That can work out too but you have to have the resources to dedicate the majority of your time. I think low turnover is much harder because you need to really understand when the margin is there. The problem for most investors i believe is they think low turnover and act high turnover, esp value people, which is just the worst of both worlds.

  19. Interesting ideas, I think you missed one essential part of understanding how this works…in investing, you invest time as well as capital. Investing is pretty unique in that you can get paid off over decades for a week’s work.

    As an example, I know a fund manager who made one decision in the early 1990s – buy Fastenal – and is still getting paid for this today. This decision was responsible for almost all outperformance through the 1990s and beyond, the business now manages $40bn.

    I am not saying low turnover is better… I would say the value of a high turnover approach is actually hugely undervalued…but that is because most people don’t understand it. High turnover works when you are exploiting small edges, you need to operate in a certain region/industry that you know very well and you need to be psychologically able to take a lot of swings and move in and out of stuff. Low turnover, I think, is harder because you need patience to do nothing for years and wait for the high margin opportunities. You also need to be able to identify these opportunities for what they are, not easy.

    The problem that most people have, I think, is that they think low turnover but act high turnover. So they have lots of new ideas every week and research a lot of different ideas but they never build up the experience and knowledge that you really need for low turnover, i.e. they invest their time poorly.

    So I kind of think there really is a broader point underlying what you say about knowing what you are good at and knowing how to invest your time to exploit this opportunity (my experience so far suggests that you can make high turnover work with very little investing talent, low turnover is very difficult without talent though).

    1. Thanks for the comment. The FAST example is a good one, and it’s one of my favorite businesses. But that is a good example to discuss, because it’s one of the best performing stocks (actually probably THE best performing stock) since 1987 when it IPO’d. I think it has averaged about 20% annually since its IPO. So if a fund manager is trying to achieve 20% annual returns, this manager would have to find the next Fastenal, and put all of his money into it. It’s unlikely that this is realistic, and it definitely wouldn’t be prudent. I think it’s a lot easier to look back at examples like WFC, MKL, BRK, FAST, Washington Post, etc… and fall in love with the idea that all you need to do is buy these types of compounders and you’ll make 15-20% annually. I think it’s harder to implement this type of an approach.

      But, I think this post has gotten somewhat misinterpreted. High turnover doesn’t mean active trading. I’m just really saying that turnover–some level of turnover–is necessary to produce the type of results that the superinvestors have produced over time. Some might not be interested in attempting to achieve these results, and there is certainly nothing wrong with owning an index fund, or owning a basket of high quality firms for decades. But my guess is if you build a portfolio of 10 or more stocks and let it sit for 20 years or more, you likely will end up with a result that will be extremely lucky to beat the index by more than 2 or 3 points. There just aren’t that many stocks that can compound intrinsically for 15% annually for decades. And stock prices tend to match intrinsic value compounding over many years.

      So I’m a huge proponent of looking for compounders, I’m just generally saying that buy and hold forever is unlikely to produce results–at the portfolio level–that will fall into the superinvestor category.

      1. Hi John

        I wanted to point out that even looking back it is really hard to find a diversified group of compounders who can give you 15% CARG year after year. Here are the “compounders” mentioned above over the last 10 years:
        (I used the Dec 31 2005 stock price and today’s price to figure the 10 year returns below)
        Ticker CARG %
        FAST 15
        MKL 7.5
        BRK.B 10
        WFC 7.5
        GHC 2

        Average 8.4% (maybe add 1% dividend since GHC pays less than 1 and BRK and MKL pay none)
        S&P 500 including Dividends 8%
        So even picking these fantastic businesses with no losers it is hard to see meaningful out performance over a ten year period. Also if we add XOM then the average would even be worse.

        So the point is just because something was a great compounder in the 80’s and 90’s doesn’t mean you can get that return to continue by choosing those stocks. If we could figure it out then we all would have put all our money in apple back in Dec 2005 and just left it there for 10 years.

        But to me, it is much easier to identify stocks that have a decent chance of doubling in the next 5 years than it is to pick stocks that will extremely out perform over 10-20 or even 30 year periods! In fact I think it is kinda crazy to try.
        So my two cents, buy undervalued stocks (fifty cent dollars) be patient. Sell when they return to full value. Wash, rinse, repeat.

        Thanks again for such a thought provoking piece.


  20. In the 50’s Buffet achieved this results because of arbitrage. It is explained in Mary Buffet book. And Peter Lynch achieved this result I think because he didn’t stop to visit companies and talk to theris CEos kwowing the business a lot. He was a very active investor. One must ask yourself, do this investment style suit me?

    Sorry for my english. Thanks for this blog from Spain!

  21. John,

    I wanted to echo the sentiments of the above comment from Seb. I, too, have a concentrated portfolio and I believe it’s important to have my point of view challenged. It’s best to understand WHO you are deeply, because eventually market cycles test your individual resolve. The worst enemy is yourself. Thus, running a concentrated portfolio allows me to maintain discipline when the markets turn. I saw this recently when numerous blog posters on other sites were “unnerved” at their “exposure” when the price of oil dropped by over 50% in late 2014. They questioned why they even owned these companies. Seemingly forgetting that these choices to purchase these securities were their own. Personally, I have found that investments with simple to definable “Moats” (like COKE’s brand & worldwide distribution system, Wells Fargo’s low cost deposit base, JNJ’s three legged platform– branded drug, consumer products, orthopedic/surgical medical devices, Markel/Berkshire’s multi-decade culture of cherishing being relied upon, Exxon’s rigid adherence to conservative ROIC decison’s on capex projects) enable me to not only maintain composure when times are challenging, but also add meaningfully during those times. Again, I just wanted to thank you for helping me challenge my thinking.

    1. Thanks for the comment Achit. Glad you found the post helpful. Yeah it’s interesting to consider all of the different tactics within and investment strategy. I think those companies you mentioned are incredible businesses, and I think each investor has to have their own approach that they are comfortable with. I think most importantly, it is crucial to think independently.

      Thanks for reading.

  22. Interesting read. Thanks for that. As many investors I have thought about this topic many times…My personal opinion based on experience is just that the high turnover and low turnover of the portfolio brings in real life a lot of practical problems.. In high turnover in order to succeed you should have be capable to find so many profitable ideas and cut the unprofitable ones very fast, so it is much harder to do than say it…With low turnover you should be very good at investment, because as consequence you will concentrate in few positions in order to get high returns with low portfolio turnover. Another issues is that Short portfolio would have to have much higher turnover than the long one, special situations portfolio much higher turnover than value portfolio.. The portfolio turnover will be highly depended on you sub-portfolios and strategies…
    Read here more about risk management and stocks valuations:

  23. John

    Thanks for the post – I enjoy reading your posts (and have sometimes commented before). I believe you make a good point here. There has been a lot of commentary here and I don’t want to go over ground you’ve already covered.

    I agree with what you are saying about how to manage small pools of capital and it is entirely consistent with Buffett has said all along. However, some things just to be aware regarding all the commentary above of include:

    1) Buffett’s 50%+ returns in the 1950s had the strongest tailwind imaginable. The 1950s were the greatest decade for stock returns with the DJIA producing 20% per annum returns on average. This chart shows you could not have better background noise than the siren song of the 1950s:

    2) Buffett’s 50%+ returns came when he was investing his own funds and/or through his partnership. Therefore, the returns would have been pre-tax returns (i.e., the taxes would have passed through to individual tax returns and would not have shown up in the mentioned 50%+ returns. Contrast that with Berkshire’s returns which are after-tax returns which have been taxed at the corporate level (usually at a 30%+ rate). Of course you might note an offsetting point: that Berkshire has likely been more levered (with float) over time than Buffett the individual or his partnerships were.

    3) Be careful of comparing after-tax returns in various decades as individual, capital gains and corporate tax rates (and actual rates levied) have changed dramatically over time. Look no further than this chart:

    1. Great points Ted. Thanks for the comment. I have a lot of follow up comments on the tax subject in general, but maybe I’ll save for another post. Thanks for reading!

  24. Fantastic post (as usual), John. This post definitely made me rethink some of my views. For those that are interested in the math, here’s a pretty eye opening article on how most people overestimate the tax benefits of low turnover strategies (a la John’s point above):

    The kicker: “An investment that changes just once a decade actually forfeits more than half of the tax deferral benefits over the span of 30 years, and for a portfolio with dividends as well, a mere 10% turnover forfeits more than 2/3rds of the tax deferral value. In a lower return environment, the true tax deferral benefit of extending the average holding period of an investment from 2 years to 5 years – chopping the portfolio turnover rate from 50% down to 20% – is actually less than 5 basis points, which can be made up in the blink of an eye through a lower cost investment change or a mere day’s worth of relative returns (not to mention weeks, months, or years)!”

  25. Dear John,

    Thanks for your effort in providing quality content. I am personally amazed that so many people don’t see the forest behind the trees. i.e. paying too much attention to taxes and trying to pull your suggested approach to differnet directions.

    I am personally, have been mentally struggling with this high/low turnover issue for many years. The reason is simple. I have too much respect for prople like Buffett/Munger, who seem to have preached a low turnover strategy. However, in my current situation (with little capital), I logically see that high turnover strategy works better for me. So it was an eye opener to come accross your srticle, which actually helped me to persuade to forge my own path in this business. So, I am grateful for your work.

    PS. As I am not a US citizen, I don’t pay taxes in US, but only in my home country (10%), so the high turnover strategy should be the correct way to invest.

    One more point, in regards to the difficulty of finding ideas for high turnover strategy. I woulk say it is easier to find ideas that will make 10% than 100%. So to make 30% on a yearly basis, it is only enough to find 3-4 ideas a year! Which should not be difficult at all. or 6-7 ideas that would give 5% return on my capital.

    Thanks again!

  26. what a great article I am heartfull thankfull for that masterpiece you have writtten down. On the other hand it is incredible to search for any piece of advice from value investing blog’s. John keep going. You have gotten another voracious reader,

    Best wishes,
    Tomasz Artur Marcin Mielniczuk

  27. Just a superb article John! I’m a long time cigar butt investor with a five year hiatus investing Buffett style. I like cigar butt investing much better. If you haven’t read it yet you should check out the book “Warren Buffett’s Ground Rules” — it gets at a lot of the points you make. Have read pretty much all of Graham and Buffett’s writings but have to say this article and the comments are probably the best, most helpful thing I’ve read on turnover, portfolio returns, and taxes. Thank you for adding some clarity to my thinking!

  28. Super smart and original, THANKS!!

    I’ve often dissected the Dupont Formula (Productivity x Profitability x Capital Structure)
    to find companies improving the right levers before the market identified with an improved ROE. This article makes a strong argument and forces me to rethink the investment portfolio process and exactly how productivity can create improvements. It makes sense to consider the Dupont formula concepts towards running an improved investment portfolio? Yes, there are many moving parts, future uncertainties, and individual goals. But investment portfolio productivity is and I often overlook. Examples of Peter Lynch and Klarman using productivity are powerful examples!
    One, well many things I struggle with are deep value ideas are often illiquid with massive spreads. It’s challenging with the goal of using a higher turnover but as you pointed out important to consider.


Leave a Reply

Your email address will not be published. Required fields are marked *