To help you understand some of the mathematical justifications, let’s use an extreme example of a long-term investor who only thinks in 25-year time horizons (if you’re perfectly average, upon becoming an adult, you’ll enjoy two of these in your lifetime as the best compounding period you can hope for under an average statistical life expectancy is a bit north of 50 years unless you’re lucky enough to enter the world with atrust fund).
Imagine you buy $100,000 worth of Acme Industries, Inc. You hold it for 25 years, it pays no dividends, and at the end of that period, it is worth $1,000,000. At this moment in time, you have a built-in unrealized gain of $900,000.
Were you to sell your position, that $900,000 would trigger Federal, state, and local taxes, depending on where you lived. Under the new tax rules (20% tax for long term positions + 3.8% special Obamacare surcharge tax on unearned investment income if you make over a certain amount in any given year), the best case scenario is you would forfeit $214,200 to the Federal government upon the liquidation of your investment. In my home state, there would be an extra $41,200 or so in taxes added on top of that, totaling $255,400. If you reside in a city such as New York, you’d get slammed harder as your local municipality would demand further confiscation. (This is one of the reasons, incidentally, that the wealthier you are, the more sense it can make to retire to a state like Florida, where you can cash in your capital gains at 0% state and local rates. It matters not one iota that you spent your lifetime building those capital gains in a higher tax state.)
The Deferred Taxes Act as an Interest-Free Loan to Your Investment Portfolio
At this moment, you have $1,000,000 worth of investment capital working for you. It consists of:
- $100,000 initial cost basis
- $644,600 net-of-taxes unrealized gains
- $255,400 in deferred taxes (economically similar to an interest-free loan that only comes due when you choose to trigger it by selling)
Imagine Acme has an earnings yield of 5.00%. You are indirectly earning $50,000 per year on your money because you have the full $1,000,000 working for you.
If you sell your stake, triggering the tax, you will have $744,600 in cash sitting in the account, waiting to be reinvested. To generate that same net look-through profit, you need to find a company with an earnings yield of 6.72%. That is 34.4% more profit, assuming growth and valuation are identical, just to break even and be in the same position you are now. Plus, you are likely at a disadvantage with the new business because you aren’t as intimately familiar with it as you are your existing, long-term holding.
Imagine that Acme had paid out 50% of its profits as dividends ($25,000 per year, for a 2.5% yield on the $1,000,000 holding). In this case, you would now need to find a business with a 3.36% dividend yield unless you’re willing to take a significant cut in passive income.
Economically, that $255,400 is a sort of interest-free loan; a topic I’ve touched on it in the past at Investing for Beginners. The implications of understanding these are not inconsequential (e.g., Vanguard founder John Bogle has written case studies of the net results owned by equity mutual funds with otherwise identical returns, one taking advantage of this tax-arbitrage and another rapidly trading stocks for the sake of moving to greener pastures; the tax-efficient portfolios crush their brethren, resulting in a roughly 2%+ per annum compounding advantage over long periods of time. Over 25 year holding horizons, the additional wealth is staggering, even owning a nearly identical selection of securities.)
That doesn’t mean you shouldn’t sell, only that the threshold for an intelligent move is much higher than most investors realize.
To put it into perspective, if you had insight into the restaurant industry, you would have been much better off a few years ago selling your Acme shares for what appeared to be richly valued Chipotle shares. The initial look-through earnings would have taken a big hit, but the growth was so rapid in those underlying profits that it wasn’t long before they had not only caught up, but blasted past, the comparable level of earnings Acme would have been generating for you. Likewise, had Acme been in the video rental industry, which was doomed with the rise of digital distribution, you would have wanted to sell regardless of current valuation as extinction was the inevitable end game.
Selling Really Great Companies Due to Slight or Even Modest Overvaluation Is Often a Mistake
When you find a really great company (high returns on equity, strong competitive advantages that give it a dominant position in its economic sphere, an industry that will still be around for a long time, and a shareholder friendly management that rewards owners with ever-increasing dividend checks), and you have big built-in capital gains, it’s often a mistake to sell the shares solely because they happen to be 20%, 30%, or even 40% overvalued. There might be other compelling reasons to sell, but overvaluation alone is often insufficient. It’s just the way the math works.
Dr. Jeremy J. Siegel made this point brilliantly a decade or two ago in the Journal of Portfolio Management. He was curious what would have happened if a buy-and-hold investor had bought the so-called “Nifty Fifty” stocks, which represented excellent businesses, at nose-bleed valuations; firms like Coca-Cola, Procter & Gamble, Merck, Johnson & Johnson, McDonald’s, General Electric, Anheuser-Busch, IBM, and Walt Disney. Despite being a case study for value investors warning against the dangers of overpaying – and the pain was significant for many, many years thereafter, which is the one of the reasons I often invoke it – over a 25 year period, the basket of excellent businesses bought at unjustifiable high valuations still beat the market by 0.7% per annum. The underlying returns on capital were powerful enough that they made up for a lot of stupidity. One interesting cause of this is the “lottery ticket” effect, as Wal-Mart played a big role in securing this ultimate result. You also had the power of dividends and spin-offs making a huge difference in cases like Eastman Kodak, where the company itself went bust.
(Of course, investors who waited just a few years had an opportunity to buy the Nifty Fifty at a fraction of their former market quotations, getting the best of both worlds – huge compounding rates plusattractive initial prices; a rare instance of having your cake and eating it, too.)
This is one of the reasons you see billionaires like Charlie Munger warning investors:
Over the long term, it’s hard for a stock to earn a much better return than the business which underlies it earns. If the business earns six percent on capital over forty years and you hold it for that forty years, you’re not going to make much different than a six percent return – even if you originally buy it at a huge discount. Conversely, if a business earns eighteen percent on capital over twenty or thirty years, even if you pay an expensive looking price, you’ll end up with one hell of a result. (Source: Poor Charlie’s Almanack, 3rd Edition, Page 206)
The unfortunate reality is that very few companies are capable of earning high returns on capital for periods of longer than a decade or two. Technological change and other market factors tend to push the world toward creative destruction. That’s what makes those that can so magical.
Once your existing holdings get up to where your unrealized capital gains exceed 100% on cost, for mega-capitalization blue chip stocks, I think the point at which it becomes very compelling to switch is when the long-term Treasury bond yield begins to exceeding the earnings yield by a factor of 2-1 (if it ever reaches 3-1, which has only happened a few times in history, run for the hills). I wouldn’t be able to sleep at night if that condition were persisting for too long.
Going back to our Acme example, that means if you could get 10% by owning Treasury bonds, taking the tax hit, and investing the $744,600 at that rate, collecting $74,460 in cash per year, might be a wise decision. While you would be taking on greater inflation risk, you’d also be collecting several times the passive income. What makes this complicated is the differing tax treatment between interest and dividends; especially for those in the top tax bracket for whom the $74,460 might be subject to effective rates approaching 50% between all three levels of government, leaving $37,230, while the $25,000 in dividends might result in a net $20,000 or so in cash. It’s still a big advantage in favor of Treasury bonds under such a hypothetical, but not nearly as large as it would seem at first glance.
An even better option under those conditions might be cash generating real estate, which has some tax shelter opportunities for the cash flow, as well as superior inflation protection. If Treasury yields were higher, it’s probable that real estate capitalization rates would be, as well. It would take an odd set of conditions for this not to be the case.
In any event, this is largely an academic discussion for you to keep in the back of your mind as you think about the nature of long-term equity ownership. You cannot ignore the tax code. Simply being cognizant of it can add an extra percentage point or two to your long-term results, generating millions of dollars in surplus wealth by the end of a lifetime for no other reason than you understood the power of compounding. The question becomes a bit more complicated when you are talking about holdings held within a tax-shelter such as a Roth IRA or SEP-IRA.
After our discussion two days ago, I was thinking about investment tax strategies, again, this morning as I sat in my living room with a notebook, a cup of black coffee, and Sungkyunkwan Scandal playing on television in the background. It’s been several years since I dove into a tax strategy known as asset positioning, but a refresher course is in order.
It seems like investors don’t pay enough attention to this discipline, and it’s almost as important as being able to analyze an income statement or balance sheet. To illustrate how vital this is, imagine you were a successful physician living in a place like Kansas City. You are statistically similar to others in your demographic and are married to another well-educated, high-earner. Between you and your spouse, your annual household income puts you in the top Federal and state tax brackets. Let’s assume that yields on high-grade corporate bonds with a 10 year maturity were at their long-term average of 7.2% (sometimes they are much higher, at the moment they are much lower because we are in an artificially-created bond bubble; reversion to the mean will happen at some point unless we were to sink into a Japanese style deflationary environment for several decades, which I think unlikely).
Now, imagine that you have $1,000,000 built up in savings. There are two alternate universes, the only difference between them is where you stored this money.
It seems like a trivial thing, right? Exact same stocks. Exact same bonds. Exact same tax brackets.
It’s not.
It will make a monumental difference in your annual net cash income and the ultimate value of your portfolio.
Investment Portfolio A
- Taxable brokerage account with $500,000 in 10-year high-grade corporate bonds yielding 7.2% = $36,000 pre-tax income – $14,256 Federal taxes – $1,368 Obamacare tax - $1,223 additional Missouri tax = $19,153 net interest income
- Tax-free retirement account with $500,000 in blue chip stocks yielding 3.0% = $15,000 net cash dividends
- Net Cash Income: $34,153 per annum
Investment Portfolio B
- Tax-free retirement account with $500,000 in 10-year high-grade corporate bonds yielding 7.2% = $36,000 net interest income
- Taxable brokerage account with $500,000 in blue chip stocks yielding 3.0% = $15,000 pre-tax income – $3,000 Federal tax – $570 Obamacare tax – $686 additional Missouri tax = $10,744 net dividend income
- Net Cash Income: $46,744 per annum
The second portfolio generates an extra $12,591 in cash per year, or 36.87% more income every twelve months, despite being identical. On top of this, you enjoy another huge advantage in that you are in a better long-term position due to the deferred tax advantage that begins to accrue as a common stock position grows in value, effectively turning your regular taxable brokerage account into a sort of quasi-tax shelter. That is not a small bonus. It can add up to significantly more wealth over long periods of time.
It seems hard to believe, but these numbers understate the past real world experience of an investor in such a scenario because tax rates have risen recently. Over the prior 15 or so years, the dividend rate was only 15%, not 20%, and there was no 3.8% Obamacare dividend tax on higher income earners. That means investors today are paying an effective Federal dividend tax rate of 23.8%, not the 15.0% that has been in place for quite awhile. A quick recalculation of the figures and state taxes shows us that up until these recent changes, an investor who had followed this approach would have enjoyed an extra $1,241 in cash every year on top of that $12,591 in extra cash, bringing the surplus to $13,832, or 40.50%.
That’s effectively free money by doing nothing more than arbitraging the tax code, following the rules for how certain streams of income are treated.
This is one of the things I mean when I say there are always intelligent things to do. With very little effort, and no additional risk, you could have juiced your passive income every year by an extra 36.87% to 40.50% per annum, depending on the tax rates in place at the time. Where else in life can you achieve that kind of payoff?
Think about how powerful it is over several decades. Imagine someone who retired and went on to live another 20 years. After accounting for compounding, the strategy would add at least an extra $250,000 to $300,000 in wealth to their estate’s balance sheet for doing practically nothing, plus they’d have huge built-in tax advantages on the common stocks that would then receive a stepped-up cost basis on their death when the shares were passed to their children and grandchildren, transforming all of those compounded gains into tax-free profit. A retiree who happened to live another 25 years and continued to make wise investments very well might be able to pass on an extra seven-figures above and beyond what he or she otherwise could have.
It’s insane that our civilization has such a screwed up tax code; where retirement security is determined, in part, by how clever one is. It seems unjust that identical investors, with identical savings, and identical assets could have such varying results based on their ability to understand the rules written by the lawyers in Congress. Nevertheless, those are the rules, so you’ll be well served by knowing them, studying them, and adjusting your own holdings to reflect the best way to maximize your after-tax earnings.
Interestingly, John Shoven did a study many years ago looking at this mathematical reality for mutual fund investors. The paradox? These rules don’t work for almost all actively managed fund investors because the portfolio managers turnover the holdings so rapidly, and / or the funds have built-in unrealized gains that get triggered during sell-offs when assets need to be liquidated to pay for redemptions. This results in a much higher tax bill than would otherwise be owed if the fund investors had held the stocks directly, with a big chunk of it coming in the form of short-term gains, which are taxed as ordinary income (just as high as interest income). It takes a lot of human error to turn what is an enormous advantage into a disadvantage, yet that’s what most people have done.
None of this might seem important given the low rate environment we are in at the moment, but, like all things, it is ephemeral. The cost of not knowing this when ordinary conditions prevail can be very high. Tuck it away in the back of your mind or file cabinet. You never know when you’ll need it.
COMMENTS:
Joshua,
In regards to Charlie Munger's comment about shareholder returns mimicking the business returns...
Is this idea based on the p/e ratio staying stagnant, and therefore a 20% incremental return on equity would lead to a 20% increase in stock price?
No. It would be best to walk you through a fictional example rather than try to explain the reasoning.
Let's imagine that you have two businesses - ABC and XYZ. Both start off earning exactly $1.00 per share.
ABC compounds earning a much higher return on capital, allowing it to grow earnings per share by 12% per annum for 25 years. XYZ is much more capital intensive, has lower returns on equity, and is able to grow its earnings per share by 6% per annum over this same period.
The market recognizes this and values ABC at 25x earnings and XYZ at 10x earnings.
At the end of 25 years, ABC will be earning $17.00 per share, while XYZ will be earning $4.29 per share.
If the valuation differential still exists, at the end of 25 years, ABC will sell for $425.00 per share, while XYZ will be $42.90 per share.
What if that valuation differential doesn't remain? Let's say the fast growth is over and ABC and XYZ both trade at 10x earnings, meaning ABC experienced significant multiple compression. In this case, ABC will still be worth $170.00 per share compared to XYZ's $42.90 per share. Thus, despite going through a severe reduction in the valuation premium applied to each dollar in earnings, ABC was a vastly superior investment because its underlying returns on capital were that much better, driving higher earnings for the long-term owner.
The danger is optimism. If ABC doesn't, in fact, grow earnings at 12% compounded per annum for 25 years, that higher price is a trap; it could result in you doing worse than XYZ. Investors as a whole are often too optimistic when it comes to predicting future earnings growth.
This is why you are almost always better off paying a fair price for a company that you are certain (or as least as certain as can be in this world) will earn high returns on capital. If, God forbid, I were to go into a coma for 50 years, I'd much rather my assets be in a collection of businesses like Brown Forman, even if they were overvalued at 27.71x earnings (far too rich, at the moment), than I would a business like Ford at 12.00x earnings. Even though I'm getting much lower earnings and dividends in the former case, over time, it's going to mercilessly crush the returns I could have captured by owning the car maker.
Why do I not, then, own any Brown Forman at the moment despite knowing this and wanting it? Opportunity cost. There have been better selections out there of businesses that are also excellent; e.g., Nestle at $67.89 per share was a more attractive use of funds or Berkshire Hathaway last year trading at barely above book value.
The reason this is almost never noticed by most investors is because turnover among investors is absurdly high. People think of "long term" investing as being a couple of years. Go find a typical 40 year old and ask how many have meaningful stakes they bought 10 or 20 years in the past. Very few - the stock appreciates slightly and they sell it for a profit. It's not a very intelligent way to behave if - if - you are dealing with one of those rare enterprises that exhibit a range of economic characteristics that make it superior to the typical firm by almost every metric.
see more
Thanks for the lengthy reply Joshua. Based on your example of the valuation differential of 27x vs 12x, according to efficient market theory, the growth or low growth would be baked into the price and discounted to the same pv, correct?
Thanks again,
p.s., if you ever have the time, I thought it might be interesting to do a post on coca cola's intrinsic value at the time buffet bought it.
I interpreted your question differently then Joshua did, your talking about p/e expansion/compression detached from the actual growth of the company (expansion in the late 90's for instance when all the p/e where high).
Well, what I was attempting to refer to, was this comment by buffett: "If the financial experience of new owners of Berkshire is merely to match the future financial experience of the company, any premium of market value over intrinsic business value that they pay must be maintained."
However, I realize now that maintaining intrinsic business value doesn't necessarily mean the p/e will stay the same. But, it certainly does seem like buffet is referring to overpaying.
In my original question, I was confusing eps growth with return on equity. What I was meaning to find out was the relationship between incremental invested earnings rates and the stock price. Mainly because Charlie references compounding machines, I believe he is referring to companies that continue to reinvest excess earnings at high rates.
Joshua,
Do mid cap companies have an easier time increasing earnings per share? Would the companies in your 25 year case studies have difficulty replicating their performance simply because of the behemoth size they have grown too?
Thanks
Scott
As a class, small and mid-caps tend to have a much easier time growing (and, again as a class, perform better over long periods of time). The largest blue chips still tend to do well because their businesses have built-in inflation hedges (e.g., if inflation is running 4%, odds are good Tide, Coke, a Hershey's bar, and Band-Aids are going to be able to support a 4% price increase), enjoy some natural growth from population increases (e.g., if Coke's market share holds steady, there are more customers being born every day), share repurchases, and spin-offs. If someone were interested totally in maximizing absolute returns and were willing to accept the possibility of extreme volatility in the event of things like another Great Depression, history would indicate the best choice might be a collection of value-based small capitalization stocks held in some sort of low-cost index fund, with dividends reinvested. Over 30, 40, 50+ years, the differential is enormous based on Ibbotson & Associate's research.
Back to big caps, though: Look at Exxon Mobil. The company itself has a hard time growing earnings (adjusted for fluctuations in the underlying commodity prices), but it buys back a lot of its own stock, increasing earnings per share each year.
So, yes, the bigger the firms get, the harder it is to keep up sky-high growth but they still tend to do very well (which is why they are big in the first place). Due to the nature of free markets, if returns ever drop low enough, you tend to see spin-offs, split-ups, divestitures, and other recapitalizations or restructurings. I'm working on a post now based on a news story that shows Sears Roebuck actually beat the market over the past couple of decades despite doing terribly because the retailer, which had incredible saturation back then, couldn't grow and instead began spinning off subsidiaries to stockholders. Those subsidiaries are now worth a lot of money.
see more
Not particularly on-topic, but...
This is the best thing I read all day; thank you! It mirrors exactly what the data shows for most millionaires in the United States - their friends and family have no idea, they live perfectly ordinary lives, they avoid debt, and let compounding do most of the work. I'm thrilled to add another real world example to my case study files!
Josh-
I would love to see you do a case study on someone like Benjamin Franklin. The Mozart read was fascinating, and made me want to learn more about history's greatest investors. Thanks for the amazing blog!
:-) I'm very happy to help!
-
Joshua,
Came across your blog purely by coincidence. Incredibly insightful, informative and concise. Where were you 30 years ago? (I spent the last 30 years making other people wealthy by running their companies.) Will re-evaluate my investment portfolio. Thank you. Maria
Joshua,
Came across your blog by coincidence. Insightful, informative and concise. Where were you 30 years ago? After reading your article, am re-evaluating my investment strategy.
Thank you. Maria
Joshua, what's your opinion on using options during trends where you feel your choice companies are not at a value you are willing to purchase? I know during the meltdown late 2008, you said you used puts to give yourself premium as well as put you into a position to buy at an attractive price. Ever do a similar method when the market is too rich for your blood?
I think the problem with this is that speculators can keep bidding prices up, despite them already being extremely high. Whereas a healthy business has an effective floor (e.g. will KO ever have a 100% dividend, if the company is still healthy?).
