Or does it?

First, because it is market-cap weighted, the biggest companies occupy too much of your portfolio if this is all you own. Nothing wrong with that but just to give you some perspective, these were the companies that occupied its top 5 slots in the year 2000: General Electric, Exxon Mobil, Pfizer, Citigroup and Cisco Systems.

Things change so diversify.

Second, S&P 500 is a large cap index. It owns the biggest 500 public-traded U.S. businesses. So when you own just this, you own only one flavor of the market.

Third, tech businesses have come to occupy too big of a share of the index lately. And many of the tech businesses trade at valuations far richer than the overall market, some justifiably so, many not so.

And the reason the market affords a much higher multiple to these growth-oriented names is because of the historically low interest rate environment we find ourselves in. And low interest rates means equally low discount rates.

But then they are also more sensitive to changes in interest rates so if and when rates rise, the value of these growth-aligned businesses will decline, oftentimes precipitously.

And if what you own in your portfolio is skewed towards these names, your portfolio will suffer. It will eventually recover but that recovery could take much longer than your patience can handle. The decade of the 2000 is a perfect example. If a portfolio of S&P 500 is all you owned, you had less money at the end of that decade than at the start. We often forget but then we have the data.

But enough with the ramble and on to some number crunching. We’ll run through some test portfolios that are exclusively invested in one type of stocks and compare that to the one that’s globally-diversified across size and value spectrum yet still 100 percent invested in stocks.

The first is an S&P 500 only portfolio. Of course, this is past performance and future performance can and will differ but one thing I can almost reliably say is that if you expect an S&P 500 only portfolio to do over the next decade what it did over the last, it’s not going to happen. It could if business profits suddenly explode to the upside due to some breakaway technological advancements but these things don’t happen that often and hence.

So this below is the best and worst-case rolling returns of an S&P 500 only portfolio over these past many decades.

And just to give you some perspective on all the things that transpired during this timeframe:

• The 1987 stock market crash.
• The Savings & Loans crisis of the late 1990s.
• Gulf War I.
• Real estate recession of the early to mid-1990s.
• The Asian financial crisis 1997-1998.
• Dot-com boom and then a bust 1995-2003.
• 9/11.
• Gulf War II.
• The housing market crash and the ensuing global financial crisis of 2008.
• The European debt crisis 2012.
• And of course the pandemic.

And embedded in between these major events are the many micro booms and bursts that happen from time to time. They will always happen because it’s systemic. It’s the nature of the markets.

This below by the way is the exact annualized performance of an S&P 500 only portfolio over these last two decades…

What stands out is how most of the gains are back-loaded. That’s because S&P 500 sucked wind the entire decade of the 2000s. All the gains are packed in the decade that just ended.

So the best time to have loaded up on large-cap U.S. stocks was in 2010, exactly the time of maximum despair for a large-cap only portfolio.

But then there are businesses that are not in the S&P 500 that are smaller in size that can be found in say the Russell 2000 index. Smaller businesses are usually risker, both in terms of price volatility and in terms of their ability to survive.

The best and worst-case rolling returns for a small-size companies only portfolio over these last many decades…

And the actual annualized return for these businesses going back 20 years…

The shocker is not that small companies earned more over time. That’s expected. Not guaranteed but expected.

The real shocker is that they made you more money with less risk than their large cap brethren (compare the 10 and 20-year rolling portfolio returns for the S&P 500 only portfolio vs. the Russell 2000).

Now spreading our wings a bit more and looking beyond our borders with international stocks with a great barometer for that being the EFA index, EFA as in the Europe, Far-East and Australasia.

The same rolling returns for an EFA-only portfolio of stocks…

And the exact annualized returns for that EFA-only portfolio over the past two decades…

So not that hot compared to say the S&P 500 or the Russell 2000 but that’s expected. International stocks have treaded water for quite some time but that’s the nature of the game. They will shine again at some point, no one knows when yet no big deal.

But now we’ll mix and match all these and bring in emerging markets, value stocks, mid-size companies etc. that exposes a portfolio to all available factors and possible outcomes. The exact portfolio is not as relevant because there are tilts and tweaks you can apply based on where you find a better bang for your buck at any given moment but assuming a portfolio that’s designed keeping first principles in mind, you won’t go wrong. And of course assuming a portfolio that you will stick with, come hell or high water.

The rolling returns first…

And the exact annualized return for that global all-stock portfolio over the last two decades…

A few condensed takeaways…

• First things first, this is diversification within a category of investments (stocks) so it’s not what a true diversified portfolio can and maybe should look like.
• Small-caps (Russell 2000) did better than their large-cap brethren (S&P 500) not only from the performance perspective but also from the perspective of delivering better risk-adjusted returns. That’s with comparing the 10 and 20-year rolling returns as well as the exact annualized performance over the last 20 years. Not what you would have expected considering the hype around S&P 500 this past many years but that’s expected considering how recency bias plays tricks on us. But then we got the data.
• International stocks sucked in this timeframe and that is and should be expected from any asset category from time to time. They say that you only know when you are truly diversified is when you always own one or two segments in a portfolio that are treading water at a given time. If everything does good or bad at the same time, you have a problem. That is to say that your portfolio should own a lot of uncorrelated investments though these days with correlation of literally everything with respect to everything else approaching one (perfectly correlated), it’s not that easy but that’s our world today.
• And when you sprinkle the ‘right’ type of investments in the ‘right’ proportion, though individually they might suck, the blended portfolio almost always overcomes that individual performance disadvantage over the very long-term. For the statistically inclined, that’s because when the variance of one investment is added to the variance of the other, the resulting combined variance of a portfolio is always lower. And hence that shows up in better portfolio outcomes, not only with what you make in returns over time but making those returns with a reduced portfolio volatility. So a win-win all around.

That’s all I have to say.

Thank you for reading.

Cover image credit – Chris F., Pexels

This and the Capital Asset Pricing Model that Dr. Sharpe subsequently developed led to him winning the Nobel Prize in economics in 1990.

The Sharpe ratio…

In English, it is the excess return of a risky investment over and above the return that a safe investment yields per unit of risk (volatility) of that risky investment. This is another way of stating that what counts is not just returns but risk-adjusted returns.

The risk-free rate of course is not constant. Many use T-bills (Treasury bonds with a maturity of less than a year) as the benchmark for the risk-free rate but that’s not always right. Stocks as we know are long-duration assets (perpetuities in theory) so if you own an all-stock portfolio, you need to match that with a comparable duration asset like say a long-term bond with a 10-year or a 30-year maturity for instance.

And if you add shorter duration bonds to that all-stock portfolio, the duration of the risk-free asset and hence its return to calculate the Sharpe ratio needs to be adjusted accordingly to match duration for duration.

But in today’s never seen before interest rate world, the risk-free rate of return difference between long and short maturity bonds is not going to make or break things. In fact, you are free to ignore duration entirely and use a flat 1 percent rate of return in lieu for the risk-free rate.

So that was a bit of a technicality but what Sharpe ratio tells us is that you better get compensated for the risks you take in your portfolio. And the higher the Sharpe ratio for a portfolio, the better designed a portfolio you have. Within reasons.

Because you can play games using leverage etc. to prop up the Sharpe ratio of a portfolio but we know the thing with leverage. When things blow up, they blow up spectacularly.

But Sharpe ratio or anything to do with long-range portfolio construction and prediction does not work with individual stocks. Or individual bonds. Or a single piece of real-estate. Or your entire angel investment portfolio.

It only works with broad-based asset classes that you can derive meaningful statistics from to build a portfolio around which then forms the core part of your financial plan (more on that later).

So someone just starting out in his career could have owned an all-stock portfolio that is globally allocated, across size and value factors. Of course an all-stock portfolio means being exposed to the full brunt of the volatility of stocks but that’s expected.

And this is what you would have had to endure over the years. We are going to run through three different portfolios with this one being the one with a lower Sharpe ratio than the other two. A lower Sharpe ratio isn’t necessarily bad but if you are comparing two likewise portfolios, you’d want to the pick one with a higher Sharpe ratio as long as you understand what’s in that portfolio.

The exact mix of this all-stock portfolio is irrelevant because you can always implement your own tweaks based on which corner of the market you think offers a better value but this is what you should expect. Or at least this is what you did get going back in time.

Now these are rolling returns which means that say for the 3-year bar in the plot above, you’ll start out in 1987 and end in 1990, then move to 1988 and end in 1991, then to 1989 and end in 1992 and so on. That way, you are not picking and choosing timeframes.

So what stands out is there was a year when this portfolio declined in value by 41 percent but then there was also a year where the portfolio gained an equivalent amount. But there was never a 5-year period where this portfolio lost you money. So that’s the perk of remaining invested for the long-term.

And depending upon how lucky or unlucky you were, the difference in returns between different 20-year periods is huge. I mean there was a 20-year band in the same 1987-2020 time-period where you did almost 12x your money (+1072 percent) in one versus just 4x your money (+323 percent) in the other.

And that worst 20-year band most likely corresponds to the last 20 years that ended in 2020 as can be seen with the exact annualized returns that you were able to achieve with this all-stock portfolio.

But this 20-year band saw a lot. Starting out of the gate was the Dot-com crash followed by the housing market crash of 2008 and then of course the pandemic. Not saying that the next 20 years can’t be worse but just saying.

So if you endured through this, you should be able to endure through anything the market throws at you except for world-ending calamity. But then, your portfolio would be the least of your worries and hence.

Now someone mid-career who had amassed a reasonable amount of money towards retirement could start to temper down on the volatility by allocating say 20 percent towards bonds. Bonds are less volatile and don’t generally yield more than stocks but the lower volatility of bonds and their inclusion in a portfolio shows up in the increase in Sharpe ratio of that portfolio. Again, not necessarily good or bad but a metric you can use to compare across similarly constructed portfolios.

And the best and worst-case returns below.

The annualized returns for the same portfolio going back in time.

Not that much different from an all-stock portfolio but that was for the last 20-years where stocks were literally cut in half twice in one decade.

Now someone nearing retirement could have increased the allocation to bonds to say 40 percent. The Sharpe ratio for this portfolio as expected goes up so if you are comparing two portfolios with a similar stock/bond mix, you’d know which one’s more efficient.

And the best and worst case performance data below…

…as well as what this portfolio did going back 20 years.

Again, the anomaly with the last 20 years shows up with stocks not doing as well as they did historically and with bonds absolutely crushing it. This is unlikely to be repeated for a portfolio with 40 percent allocation to bonds over the next 20 years though.

So what kind of portfolio should you own? But before that, a bit on what I do with my (our family’s) money.

We use a core and explore approach to how we deploy our savings. The core is the can’t miss, can’t fail segment of our money that must be there when we need it and hence is invested accordingly. That makes up about 90 percent of all the money we have though that percent allocation is higher now for reasons I’ll elaborate on more below. And it’s invested in an all-stock portfolio like the one shown above for three reasons:

• 4 percent is what is typically used as a safe withdrawal rate from a portfolio to live on during retirement though that might need some adjustments considering the interest rate environment we find ourselves in. But a 4 percent withdrawal rate means your portfolio cannot afford a lot of volatility and hence bonds become an important component of that portfolio. But if say a 2 percent withdrawal rate is plenty to live on during retirement then there’s no need to add bonds as dividend income alone can fulfil your income needs and that is where we expect to find ourselves at.
• There is unlikely to be a repeat of bond market performance of the last 40 years over the next 40. So if you don’t need bonds, you should not need bonds. Yes, the right kind of bonds can and do make the ride smoother but if you don’t care about the ups and downs, you don’t need bonds either.
• And though an all-stock portfolio will have an inferior Sharpe ratio than a portfolio with a decent allocation to bonds, that in and off itself is not necessarily bad. Because there is no guarantee that adding bonds will enhance the volatility reduction benefits bonds provided in the past. So though Sharpe ratios are important to compare two similar portfolios, that is where that comparison stops. Just because a portfolio’s Sharpe ratio is lower does not automatically imply inferiority.

I also do a bit of exploring with our money and that’s where the tiny explore portion of our money is invested. We haven’t done much to this in the last many years due to the valuation environment we find ourselves in but we will at some point again when wonderful businesses could be had at reasonable valuations when this current cycle turns. And turn it will.

But of course there is no need to do the explore if you are not meeting your plan goals. And it also comes down to whether you enjoy doing all the work needed to explore because work it does take.

Plus since statistical calculations can’t be done with the explore segment of my portfolio, Sharpe ratios and things like that goes out the window. This is an attempt to eke out a bit more than what the core-only portfolio can deliver but of course there are no guarantees.

The explore portion of our money is currently invested in a bunch of businesses that are small, cash flow rich with predictable business models. At least, businesses that I can do some modeling and projections on. Businesses like Raven Industries that make precision agriculture products and engineered films. It’s been in the news lately as it is getting acquired in an all cash deal.

Other companies that were acquired since the time I first built this portfolio almost a decade back…

• Pall Corp., a maker of water filtration systems was acquired by Danaher Corp. August 28, 2015 was the last trading day for the stock.
• Mead Johnson Nutrition Co., maker of infant formula such as Enfamil brand, was acquired by Reckitt Benckiser Group. June 14, 2017 was the last trading day for the stock.
• Clarcor, a maker of filters for automotive and heavy industrial applications was acquired by Parker-Hannifin. February 27, 2017 was the last trading day for the stock.
• Kaydon Corp., a maker of industrial bearings and shock absorber systems was acquired by SKF. October 15, 2013 was the last trading day for the stock.
• Bio-Reference Labs, a provider of clinical laboratory testing services for the detection, diagnosis, evaluation, monitoring, and treatment of diseases in the United States was acquired by Opko Health. August 19, 2015 was the last trading day for the stock.
• Sigma-Aldrich, a company that develops, manufactures, purchases and distributes a range of biochemical and organic chemical products, kits and services that are used in scientific research was acquired by Merck. November 17, 2015 was the last trading day for the stock.

Then there are business like International Flavors and Fragrances. And W.W. Grainger. And Copart. And C.H. Robinson Worldwide and a few others that we’ll continue to own for a long time.

Hope this helps.

Thank you for reading.

Cover image credit – RF Studio, Pexels

Craig L. Israelsen, professor of all things financial planning at the Brigham Young University likens the art of portfolio construction to making salsa. I mean you can add your own flavor to a recipe but ultimately, the ingredients that constitute a wholesome salsa don’t differ much. And so shouldn’t the ingredients of your portfolio.

The recipe is your portfolio’s asset allocation. There are complex elements to it but inherently, not that complex.

Some of the elements that make up a salsa, say salt for example, may not be very exciting. And they are not supposed to be. They are neutralizers.

A well-designed portfolio should have its own set of neutralizers.

But then we’ve got a world where the only thing that matters is how you compare against the S&P 500 index. At least lately. Or God forbid, the Nasdaq. Talking about Nasdaq…

Comparisons against these indices didn’t happen as much during the decade of the 2000s when indices similar to these sucked wind but they happen now because we forget. We are too busy. We take mental shortcuts instead of thinking deeply about what we own and why.

And since when do we compare salsa to 500 ground-up tomatoes? We have created a misconception around what diversification is. The S&P 500 is a diversified set within an asset class. That’s not true diversification. That’s intra-diversification. It’s depth, not breadth.

And Nasdaq’s worse. Yes, yes, a lot of companies that make up that index will go on to change our world but it’s still mostly a narrow subset of the investible universe. And there are other means to own it.

So getting both breadth and depth is true diversification. A multi-asset portfolio that encompasses a need appropriate allocation is what you want.

A 60-year old’s portfolio should look different than that of a 30-year old.

A 60-year old’s portfolio with a solid pension should also look different than that of a 60-year old’s without a pension.

Back to the S&P 500, this is what you get when you own just it.

So the biggest 25 businesses take up almost 45% of your money. Even in a winner take all kind of an economy, that’s too much concentration in just a few businesses if this is all you’ll do with your money.

And large businesses don’t stay large forever. They become stodgy, bureaucratic and unmanageable and eventually get replaced by smaller, more nimble rivals. It is only a rare breed that can maintain their market power for decades on. A prime example of that process and there are many is General Electric which at one point was the largest market value business in the world but now is a shell of its former self.

The current bunch of the large cap universe could be an exception but history says otherwise. There is always something around the corner that would dislodge the hot ones of today. It might take longer but it’s going to happen.

Plus a portfolio invested only in the S&P 500 index leaves out stalwarts like these and 3,000+ others…

Not a collection that screams of deep value but it is a collection.

So what’s a flavor of depth and breadth? Maybe this…

That’s for domestic equities. The ingredients and the proportions can change based on where you are in life but seems even-keeled and at least pointing in the right direction. You do something similar for international equities. Then you bring in bonds and cash, real estate and alternatives. Now that’s salsa.

On bonds, the last 40 years have been the best 40 years in the history of the fixed-income world. That’s not going to be repeated so caution on what type and duration to own is warranted.

And the portfolio you design should be a function of how much you’ve saved, what you’ll continue to save plus growth in value. Expecting just your portfolio to do the heavy-lifting without a pitch from your savings is not going to cut it in this yield-starved world. Encounter anything that promises that, run.

And last, as Carl Richards, the author of The Behavior Gap says, you can have the best portfolio ever designed in the history of the world and you make one behavioral mistake a decade, you might as well have stuffed all that cash in a mattress.

So don’t.

Thank you for reading.

Until later.

Cover image credit – Karolina Grabowska, Pexels

So a very clean bell-shaped distribution.

But instead of say having access to 1 million data points, what if we had only 100 data points? The underlying population is still normally distributed but because of the smaller sample size, the distribution might not quite look bell-shaped which is what we were expecting.

It could look something like this…

So that’s the difference between dealing with sample data vs. population data. You could be dealt any hand possible based on the size of the sample you collect and the differences between sample to sample can sometimes be big. A given sample could precisely represent the population or could be a completely different distribution altogether even though the underlying population distribution is still normal.

So now that we got that straight, what does that have to do with our money? The plot below is the performance distribution of the two dominant asset classes that we have reliable data on over the past 90 years. Stocks here are represented by the S&P 500 index and bonds by the 10-year Treasury bonds. You could replace S&P 500 with a diversified global stock portfolio and you’d get more or less the same distribution but because global stock market data going that far back is not readily available, we use S&P 500 as a proxy.

Stocks are more volatile than bonds. We know that and we can see that from the comparison of the spread between stocks and bonds. And I don’t know about you but neither of these distributions look bell-shaped to me. That’s because though we have data on 90-years of performance for these two asset classes, they still represent a small subset of the population of all the returns that have happened before and all the returns that will materialize in the future. So the underlying distribution of the stock and bond market returns could still be normal and if we assume that, here’s what the distribution of returns could look like for the two asset classes over a span of say 1,000 years .

Almost bell-shaped. So the underlying distribution can be assumed to be normal even though the 90-year sample does not look anywhere close to normal. And unfortunately, we have to make that assumption to assess the impact of these statistics on our portfolios. And our lives.

So now that we got that straight, say you are a young whippersnapper with a 40-year investment time horizon. Your time horizon in reality is much longer than that but let’s just stay with this for now. You have some cash that you’ve saved up and you want to plunk that down in a portfolio of stocks that yields on average say 7% during that 40-year time frame. Why stocks? Because you are young and you can afford any level of volatility (these words could come back to haunt you) the market throws at you. And 7% for an all-stock portfolio is lower than what the markets have yielded historically but we know the valuation and the interest rate drill and hence 7% sounds about right as an assumption.

So what would you have in 40 years if a single $1,000 were left to compound at 7%. $14,974 or rounding that off to say $15,000. So that’s 15x your money.

What’s missing? That 15x assumes a constant 7% return each and every year. That of course is not real. Markets don’t move in averages. They can fall a few years in a row, then be up a few years and so on. That’s volatility and the difference with different levels of volatility on the final accumulated wealth can be yuge.

To prove that, we simulate by drawing 10,000 samples of 40-year interval from a population of portfolio returns that is normally distributed but with varying levels of volatility.

And here are the results starting from the worst-case (losing your shirt) to the best-case (making a killing).

So with a 50% volatility, 85% of the portfolios lose money over this 40-year time frame. This is akin to a more venture type of investing and you might want to do that with some portion of your portfolio but not with your entire portfolio.

Or this could also be an outcome of a very concentrated portfolio of stocks.

Historically, the standard deviation (volatility) associated with a broadly diversified stock portfolio is around 20%. With bonds, it’s about 6%. You mix the two and you really have to try hard to push the volatility beyond 15%. So in theory, you almost never lose money with that portfolio. Or at least you didn’t historically.

But avoiding capital loss is not your only goal. You are doing all this to also make some money because you can be a wage-slave for only so long.

So some more data on just how much are you able to grow your wealth by with different levels of volatility but with the same average return.

One thing is clear. Off the chart volatility kills as is evident by the 50% volatility mark in all the plots. You lose most of the time. Plus the probability of you making a killing are so infinitesimally small that you’d rather not try. But then this is where a collection of most small businesses and start-ups lie so not trying is also not good for you and me and the economy. It is these risk takers and investors willing to fund these ideas and businesses that creates this quality of life we take for granted. So we take those chances and we should but with a small portion of our portfolios.

And you don’t have to take crazy risks to do well over time. A portfolio with 5% volatility takes you quite far almost all the time. Yes, you are less likely to make a killing as shown by the ‘no bar’ in the ‘more than 50x plot’ above but you are likely to always match and exceed inflation.

And sometimes that’s all you need, especially during retirement.

Plus any portfolio with volatility less than 15% always made you money and at most times, a lot of money.

And if you are dollar-cost averaging through this 40-year investment timeframe, even an all stock portfolio works, especially during the early phase of your accumulation cycle.

But what you haven’t asked and what you should be asking is, who would go for portfolios with crazy volatility yet only earn 7% returns on average? We know the risk-return trade-off. The more risk you take, the more in terms of a return you should expect.

So then we analyze situations where returns are different – lower returns for a less volatile portfolio and higher returns for the more volatile one. That is, you are getting compensated for taking those crazy risks.

But do you really get compensated for taking those risks in the long run? Back to the data again…

The situation improves a bit for the crazy volatile portfolio but you still end up losing your invested capital 60% of the time.

So the probability of you making a killing with a highly volatile portfolio improved but not by much. In fact, a vast majority of portfolios that return 5% with 10% volatility do better that the portfolios that return 15% but with 50% volatility.

So the moral of the story is to shoot for decent returns that will allow you to meet your goals but ignore volatility at your own peril. And in fact, if given a choice between a higher return but a more volatile portfolio compared to a lower return, less volatile portfolio, choose the latter. Not only will you sleep easy but you’ll sleep easy while getting rich.

Until later.

Cover image credit – Paulo Valdivieso, Flickr

When outcome is essential and matters more than relationship, use trust, but verify. When relationship matters more than any single outcome, don’t use it. 

So true that. But since there are no relationships to care for when it comes to being a good steward of our money, we should trust little and verify a lot. And that task becomes a lot easier with a bit of intuition and access to data.

So there is this thing called rebalancing and it means exactly as it sounds. Say we start with a 60/40 stock-bond portfolio and over time as the markets evolve, that allocation drifts to say 65/35. If the intent is to maintain a constant allocation, we’d sell the stock component of the portfolio, taking out the excess and buy into the bond portion to bring the allocation back to 60/40. That’s rebalancing.

And it sounds like a great idea. We sell something that has gone up and buy the other that has gone down – a classic buy low, sell high approach. So that’s great. But then I read something along the lines that rebalancing between stocks and bonds works but rebalancing within a category is not ideal or does not work as effectively.

Take stocks as a category for example. You’ll likely own some large company stocks, mid-size company stocks and some small ones. And then you’ll own developed market international stocks and emerging market stocks and so on. Not all of them will move in the same direction all the time. Some will zig while others zag. Or some will zig some while others will zig more and so on. So if you rebalance within a category, you would be theoretically doing the same thing that you ideally would want to do – buy low and sell high.

But now there’s this doubt and we need to get to the bottom of it to make sure all’s okay. So as we would do and as we should do, we test both approaches at once to validate that what we’ve done all along was not inferior to what we should have done.

So we go back to data and test whether an invest & forget approach works better than say rebalancing annually. Ideally we would and should rebalance as often as there is an opportunity to rebalance but let’s just assume we do it once each year. We’ll use data on annual returns for four asset classes to test the two approaches.

• Large company U.S. stocks
• Small company U.S. stocks
• International stocks
• U.S. bonds

We invest $100 in a portfolio comprised of these four asset classes at the start of the entire time period in varying proportions of 10% increments and assess whether rebalancing does what it is supposed to do. A snapshot of different portfolio combinations is shown below.

Each row is one portfolio and with 10% incremental allocation spread across four asset classes means 258 different portfolio combinations we get to try this on.

Starting with year 1, in year 2 in case of the rebalancing approach, we sell whatever has deviated to the upside from the original allocation and buy what has declined to bring the allocation back in line. For the invest & forget approach, we split and invest the original $100 into the allocation we started out with and let the money ride till the end of the period. The end of the period by the way is 2018 and the dataset contains 49 years of data starting in 1970. So we are comparing the ending values of each portfolio at the end of 2018 to test the rebalancing vs. invest & forget approach.

The first thing we should do to get a good feel is to look at how the ending values are distributed between the two options.

So clearly rebalancing works as is evident from a slight right shift of its distribution as compared to invest & forget. The spread is a bit wider though with rebalancing which is not desired but a bigger question is, are we comparing the same portfolios when comparing outcomes between the two? What we should ideally compare is the ending value of portfolio 1 in the invest & forget case with the ending value of portfolio 1 in the rebalanced case, the ending value of portfolio 2 in the invest & forget case with the ending value of portfolio 2 in the rebalanced case and so on.

So that’s what we have done next and this is what we find when we do a portfolio by portfolio comparison of the ending values…

• Out of 258 portfolios, each with a different asset allocation, the ending values of 243 portfolios that were annually rebalanced equaled or outperformed those of the invest & forget ones. So a 94% outperformance rate if the portfolios were rebalanced as compared to invest & forget.
• 79 portfolios out of 258 that were annually rebalanced outperformed invest & forget ones by more than 10%.
• And the ending values of three out of 258 portfolios outperformed invest & forget by more than 25%.

So rebalancing works or at least worked almost all the time. But what if the bond allocation was held constant at say 40%? The original thesis was that rebalancing is more effective between categories (stocks vs. bonds) versus within categories (within stocks or within bonds). So trying that out…

Apparently the same story here with the shift in distribution for the rebalanced case more to the right than for the invest & forget approach. Oh and by the way, because the bond allocation is held constant with only the remaining three asset classes in the stock category allowed to vary, only 60 portfolio combinations are possible.

A portfolio by portfolio comparison of the ending values yields the following results…

• Out of 60 possible portfolios, each with a different asset allocation and a fixed bond allocation, the ending values of 59 portfolios that were annually rebalanced equaled or outperformed those of the invest & forget ones. So a 98% hit rate making the case even stronger for the rebalancing approach.
• 25 portfolios out of 60 that were annually rebalanced outperformed invest & forget ones by more than 10%.
• And one outperformed invest & forget by more than 25%.

So if you had to wager, rebalancing still wins.

What if you owned an all-stock portfolio? Would rebalancing still outperform invest & forget?

Appears to be a yes. And again as before, only 60 portfolio combinations are possible so a portfolio by portfolio comparison yields the following…

• Out of 60 possible all-stock portfolios, the ending values with the rebalanced approach equaled or outperformed invest & forget each and every time. So a 100% hit rate in favor of rebalancing.
• But none of them outperformed by more than 10% so not a big thumping vote for one over the other.

But what if the returns of the past do not repeat in the same sequence? Could the outcomes be different with a different sequence of returns?

To assess that, we sample returns for each asset class randomly and recreate the asset class returns dataset each time and compare the ending portfolio values between the two approaches. And just to make sure that we have at least attempted to try every which way to convincingly make one approach fail over the other, we do this 500 times. The results…

With portfolios constructed out of a combination of the four asset classes (large company U.S. stocks, small company U.S. stocks, international stocks and U.S. bonds)…

So a very strong vote in favor of rebalancing even with randomized returns sequences.

With a 40% constant allocation to U.S. bonds and the allocation to stocks allowed to vary…

Rebalancing wins here as well.

And for the stocks only portfolios (large company U.S. stocks, small company U.S. stocks & international stocks)…

So you’d be crazy to not rebalance your portfolios from time to time.

But here’s a thing. This whole thing is fundamentally based on the fact that mean reversion will always happen. That is, if an investment has deviated from its normal course either on the upside or the downside, it will always revert back to its mean course over the long term.

But what is long term? Ten years, twenty-five years, hundred years? We can only know this in hindsight maybe long after we are dead so that’s one thing to consider.

And what if an investment ceases to exist? Individual companies we know live and die all the time so to guard against that risk, we’d diversify into a sector. Could an entire sector vanish or never, ever revert back to its mean trajectory of growth? Of course.

What about countries? That’s easy, Japan.

The post-war rebuilding which eventually culminated into a real estate led economic boom of the 1980’s Japan was so big and went on for so long that just the fact that it all eventually came crashing down does not quite do enough justice to the sheer scale of that bubble. Edward Chancellor in his book, Devil Take The Hindmost chronicles the reasons for the boom and what led to its eventual implosion.

One of the key drivers for the boom…

Between 1956 and 1986, land prices increased 5,000 percent, while consumer prices merely doubled. During this period, in only one year (1974) did land prices decline. Acting on the belief that land prices would never fall again, Japanese banks provided loans against the collateral of land rather than cash flows.

Land prices will never fall again, wonder where we have heard that before? So the banks lent money just because the value of the land rose. And the more it rose, the more they lent, creating that self-fulfilling feedback loop of ever increasing prices, leveraged to the hilt. Things got so crazy that by 1989,

The grounds of the Imperial Palace in Tokyo were estimated to be worth more than the entire real estate value of California (or Canada, if you preferred).

And the post-crash recovery didn’t quite materialize or hasn’t yet materialized due to structural reasons that are unique to Japan, though there are signs that things might be finally on the mend. But then they have a long way to go.

Back to the rebalancing or not rebalancing question at hand, so if a portfolio design is done not considering the fact that there might not ever be a mean reversion, we are doomed.

And I might have insinuated before that I am strongly in one camp or the other but I am not completely sold on either. So I employ a mix of both. And that’s because I don’t know the future. No one knows the future but try one must with as much supporting research and evidence. And a bit of intuition.

Thank you for reading and persevering through.

Until later.

Cover image credit – Matthew T Rader, Pexels

So what’s the paper about? Designing diversified portfolios through the use of uncorrelated asset classes as a way to invest optimally with a goal to earn the highest rate of return for a given level of risk. Yes, it’s a mouthful but this simple yet intuitive approach to portfolio management didn’t exist until 1952. At least no one attempted to formalize it and present it in a way he did.

And the ‘right’ amount of diversification is all about correlation or to be more precise, uncorrelation between different investments in a portfolio. Owning stakes in 10 different technology companies is not diversification. Or spreading your bets between say a portfolio that invests in the Dow vs. the S&P 500 index is not diversification. How do we know? Because again, correlation but before we jump into that, a bit about the different types of investments or asset classes you could consider to design your portfolio.

So we know an investment. Buying that stock in the hope that it appreciates in value while it pays dividends (or not) is an investment. Buying that bond that pays interest until it matures is an investment. Or that piece of real estate that provides rental income is an investment. But what’s an asset class? I got my hands on returns data published in the Feb 2019 issue of the Financial Planning magazine by the Steele Mutual Fund Expert for 7 major asset classes going back to 1970. We’ll use this and probably some other data series to do whatever we need to do here.

Those asset classes are Large U.S. Equity, Small U.S. Equity, Non U.S. Equity, Bonds, Cash, REITs and Commodities.

Large U.S. Equity comprises of all large publicly traded companies in these United States. What constitutes large could be different based on the organization assembling that asset class but in general, it means all companies with market capitalization of $10 billion or more. Then there’s an asset class that owns small companies. There’s one that owns non-U.S. companies, there’s one that owns bonds, real estate etc.

Real estate here is not the home you own. Owning that is akin to owning a single company stock or a bond. Real estate here is the entire real estate sector represented by REITs or Real Estate Investment Trusts. What’s a REIT? It’s a publicly traded entity that owns or finances income producing real estate spanning a variety of sectors. For example, that mall you just visited, that could be owned by a REIT. Or that apartment complex where you rent your home? That’s more than likely owned by a REIT. So is that office park or that hospital etc. So a REIT might specialize in one type of real estate but REITs as an asset class gives you broad exposure to every type of real estate. That’s diversification but diversification within a segment of an economy.

And commodities is exactly what it means, owning commodities like gold, silver, copper, oil, wheat, corn etc. You don’t own them physically but you own a stake in a collection of them and someone holds and keeps them safe for you. Yes, you have questions and I’ve got to explain more but for now, just assume that you bought a collection of stuff and held it till 2018 (the last year for which we have data for).

But why not do all this portfolio optimization with individual stocks? Or say bonds? The math is the same so it is doable but what we have here is 49 years of data. And we know a lot can change in 49 years. How much? Take the Dow Jones Industrial Average (DJIA) index for example and compare its constituents in 1970 to the present day constituents.

Only 4 companies that existed in 1970 still remain in the Dow. What happened to the rest? Some got acquired or merged with other companies but most flamed out. And yet the DJIA continued to march on higher from around 800 points in 1970 to 26,000 points today. And that’s not factoring in any dividends. So if you’d done this math in 1970 with the companies of that time and built a portfolio around them, your situation is likely not looking that hot now. And building statistics-based portfolio models requires that the asset class does not vanish which it does in many cases when you build a portfolio around just individual stocks and bonds.

Okay, so we have returns data on these seven asset classes. How can we get a sense of how they have performed over these many decades? We start with the distributions. What do they look like? Are they narrow or wide? Is there any bimodality in the data? All this with this one plot below.

Stacking the plots makes comparisons easy and here’s what we can conclude…

• That faint line in the middle of each plot is the median. Half the returns are below that line and half above. From what I see, small U.S. equity seems to have the highest median returns followed by real estate and then large U.S. equity. I say median returns for large U.S. equity and real estate are almost identical. Median returns for commodities and non-U.S. equity are comparable which are then followed by bonds and then cash. So we should know which one would have generated the most amount of wealth, right? Small U.S. equity for now but we’ll see.
• As long as the returns are randomly distributed, you’d want more of them to be on the right of the 0% line than to the left which happens to be the case with each asset class above.
• Cash and bonds have lower spreads (risk) and lower median returns than other asset classes. That’s expected if you believe that there should exist a risk premium as you take on more volatility with your investments.
• The spread (standard deviation to be more precise) for commodities appears to be the widest but quite a big chunk of the returns happens to be on the left of the zero line when compared to other asset classes. That has implications and we’ll see.

So what made the most amount of money?

Real estate or REITs to be more precise and by a wide margin. And look at the difference in value between that and say small U.S. equity. Almost a double in REITs vs. small U.S. equity even though the median return for small U.S. equity is in fact higher than that for REITs. So why this apparent discrepancy?

You would have sensed it by looking at the spreads but variance or the volatility in returns is what makes that big of a difference. Not that REITs didn’t have years where the returns were negative but they were not as many as small U.S. equity. And look at commodities. You would have made more money being an investor in supposedly safe bonds than in commodities even though there were more years with higher returns in commodities than they were for bonds.

Before we move on, a bit about the median and the mean (average). Median is the half way point when you sort a data series in ascending or descending order. Say you have a data series with 5 data points; 3, 2, 5, 9, 7. The sorted series then is 2, 3, 5, 7, 9. The median hence is the number 5. We know the average or the mean and that is (3 + 2 + 5 + 9 + 7) / 5 = 5.2. Had to get this out of the way because up until now, we have been making statements using the median values but we need to come back to the mean because that is what this is all about.

The summary then for the average or mean returns and volatility for the 7 asset classes under consideration is as shown below.

REITs actually earned just a hair bit higher annual returns on average than small U.S. equities but with 3% lower standard deviation. And an investment in commodities sucked even after earning 9.5% returns on average and thank the volatility number associated with that asset class for that.

So just returns are not enough. Risk adjusted returns is what counts. A 50% drop in the value of your portfolio does not take a 50% return back to break-even. You need a 100% return to get back to what you started out with. So minimizing that drop in the first place means that it would be a lot less harder to come back to where you were in case your portfolio experiences bouts of volatility. Which it will from time to time.

But even investing in the ‘best’ of asset classes did not come without its own issues. Compare for example bonds to large U.S. equities and REITs.

Bonds of course didn’t make you as much money but they allowed you to sleep like a baby as is evident from the drops in value above from time to time for large U.S. equity and REITs as compared to bonds. Another way to calculate the frequency of heart burns you’d have to endure is to compare drawdowns between the three asset classes.

So quite a few times, you experienced gut-wrenching drops in the value of your portfolio in REITs and in large U.S. equity even though in the end, you came out way ahead. And there’s no guarantee of anything. That 50% drop could have turned into a 60% or a 70% drop. That’s the price you paid to make all that money by persevering and hanging on through that for dear life. And that’s if you did but not many do.

If you cannot handle this extent of volatility, you add bonds and cash because as you can see, there’s hardly any volatility associated with either of them. You didn’t make a killing but as stated before, you slept well. But there’s a caveat especially with bonds which we’ll get to later.

So how big of a slice should bonds and cash occupy your portfolio? Or better yet, how can you create a portfolio that lets you choose the amount of heart burn you are willing to endure? And what’s the ideal portfolio mix that gets you the best return with the least amount of risk? We use mean-variance optimization (or as Fredo would say, “I’m smart.”) to attempt to answer such questions.

Say you mix and match different investments in varying proportions and you get portfolios with risk-return scenarios like below.

Risk here of course means volatility (standard deviation). One of those portfolios is marked in red and the other in green. Both delivered the same return but with starkly different risk levels. And of course you’d pick the lower risk portfolio for the same given return.

Or how about the two portfolios below in red and green?

You’d not pick a portfolio that’s red over say green. Why would you.

Or say you are 22, just out of college and in your first job trying to decide what investments to populate your 401(k) with. You could and should decide to go all out on the risk-return spectrum by choosing a portfolio shown in green on the far right below. You don’t quite yet have as much financial capital to worry about volatility but you sure do have plenty of human capital ahead of you that you’d slowly and eventually convert to financial capital. And because you are adding to your savings with each paycheck, a bit of volatility might actually help than hurt as you get more opportunities to accumulate assets at depressed prices.

Or you could be in retirement where you have pretty much exhausted your human capital and are sitting on a boatload of financial capital that you would slowly extract to live on. You’d rather then own the portfolio in green shown on the lower left.

You would have sort of noticed a theoretical upper bound across the risk spectrum in terms of the returns you can expect from combining investments in varying proportions as shown below.

That’s what’s called the Efficient Frontier. Portfolios on that frontier are considered optimal, offering the highest expected return for a given risk. Portfolios that lie below that frontier are considered sub-optimal and do not generally compensate for the portfolio risk you bear.

So now that we’ve got that straight, we’ll use the data we have on those asset classes and assign them weights in 10% increment and create portfolios (7,658 of them in total) to see which ones lie where on this risk-return spectrum.

We’ll pick 5 different portfolios to dig a bit deeper into their contents and to extract any insights if any.

Portfolio_7658 is the highest risk portfolio that’s allocated to and you guessed that right, 100% into commodities. And it’s not an optimal portfolio because it’s nowhere close to the Efficient Frontier.

Portfolio_920 is the lowest risk one that owns 10% bonds and 90% cash. No surprise there.

Portfolio_2823 delivered the highest average return and it’s comprised entirely of REITs. Again, expected as we saw before.

Portfolio_4575 lies on the Efficient Frontier with 10% volatility. It’s allocated to 20% large U.S. equity, 30% to bonds, 40% to REITs and 10% to commodities. A bit decent but not ideal and will explain why (I am not done yet ).

Portfolio_4755 lies on the Efficient Frontier as well but with 15% volatility. It’s allocated to 20% large U.S. equity, 70% to REITs and 10% to commodities. Again not ideal.

So now you start to see issues with formulaic approach to portfolio construction. Some don’t make sense, some are too heavy into a few asset classes and some are overly lop-sided. But just for the fun of it, we’ll see what each of these portfolios did if you’d picked one of them at the start of the period and rebalanced annually to the same allocation you started out with.

As expected, Portfolio_2823 did the best as it was entirely comprised of REITs. Portfolio_4755 did about the same and was more diversified. I would have picked that over Portfolio_2823 any day though that still was REITs heavy.

But the entire premise of all this is that it relies on historical data. You can do all the math you want but we know that thing we hear everywhere we look. And that is, past performance is not a predictor of future results. How would you have known to pick only REITs when you created that portfolio 50 years ago? That’s your entire adult life. You’ve got this one shot to get from point A to point B so taking that chance requires a level of obliviousness that borders on well, obliviousness. Or even if you did pick it ‘right’, what are the chances that you hung on through all the ups and downs that a heavily concentrated portfolio would have exposed you to.

Plus what happened in the past is unlikely to be repeated again, at least not in the same way and that’s all due to what interest rates have done over these last many decades. Take bonds for example.

We know the deal with bonds. As interest rates go down, bond prices go up. And up they have with the relentless bull market in bonds since the early eighties when interest rates were double digits to where they are now. Can the bond bull market continue? Not a chance and hence building portfolios based on historical returns data on bonds will of course not turn out great.

That interest rate tailwind is there for stocks as well.

Why? Say you need to make a capital investment to increase production of whatever stuff you are in the business of making. So you go to a bank for a loan for say a duration of 10 years. The prevailing interest rate at the time is say 10%. Now you make those interest payments on time for the first year and record whatever profits your business generated which of course is net of interest expense. But say the rate declined to 8%. What would you do as a steward of that business? You’d run to the bank to refinance at the now lower rate. You suddenly don’t have as much interest expense and hence your net profit rises. And so does the value of your business or the stock price, all else remaining constant.

And it’s even truer with real estate than with stocks.

Real estate is packaged commodities. It just sits there. It does provide a service and that is shelter but beyond that, not much. It’s not going to create a cure for cancer or reinvent the way how we live or travel or communicate. It’s also an extremely interest rate sensitive asset and likely more so. We see evidence of that with the obvious negative correlation above between rates and an investment in REITs. Can the REIT out-performance continue? Very unlikely.

But it’s not that you completely ignore this theory. You use a bit of it and a bit of your understanding of history, business and the economy and create a portfolio that is just right for you. I’ve shared some of my thoughts on how to go about doing that here and will do more from time to time. But in the end, you are the one who will have to persevere and endure all the ups and downs associated with your choices in the coming decades. Because as Morgan Housel quotes,

Something stupid you can stick with will probably outperform something smart that you’ll burn out on.

So your ability to stick with what you own in your portfolio provided you have justifiable (to you for sure) and quantifiable reasons to own what you own is what will ultimately count.

So long and long.

Until later.

Cover image credit – Quang Nguyen Vinh, Pexels

So that’s that but the other big risk of course is the probability of outliving our savings. And volatility is one reason we tune our portfolios depending upon when, how much and for how long we are going to draw income from our savings.

But what could really throw a monkey-wrench into all our planning is the pattern associated with that volatility. Are the ups and downs of our portfolios random or is there an inherent pattern to how those returns transpire? Peter L. Bernstein in his book ‘Against The Gods: The Remarkable Story of Risk‘ hints at caution against over-reliance on historical data for just this reason.

So we pour in data from the past to fuel the decision-making mechanisms created by our models, be they linear or nonlinear. But therein lies the logician’s trap: past data from real life constitute a sequence of events rather than a set of independent observations, which is what the laws of probability demand. Even though many economic and financial variables fall into distributions that approximate a bell curve, the picture is never perfect. It is in those outliers and imperfections that the wildness lurks.

That is that even if you were able to extract the key statistical measures from past data, you don’t want to just go on and use them to predict what future returns will amount to. That’s because without taking into consideration the current market environment and those outlier events, you could be way off from what you planned.

And the sequence of how those portfolio returns come about during retirement could be one of those outlier events that determines whether you depart with millions left behind or your money departs way before you depart. The technical term for this and as implied is the sequence of returns risk. That’s one more risk we need to plan for and to understand its implication, we’ll walk through just what could go wrong even when we have done everything right.

So even though an over-reliance on past data could lead us astray with our plans, we still need some basic metrics to anticipate what future returns could look like. So we do just that and use historical performance data for both stocks and bonds in some combination of each in a given portfolio to simulate those outlier scenarios. We first fit a simulated distribution (in yellow) of returns for stocks over a distribution of the actual past returns (in pink) as shown below to see if a bell-shaped distribution of returns assumption holds.

It does to some extent for this one sample so we go with this assumption and extract the mean (average) and the volatility (standard deviation) associated with that distribution as a baseline to help predict future returns. The average historical return for stocks by the way is the black, dashed line which shows that stocks on average have returned around 10% annually during this entire time-frame.

We do the same for bonds and extract the relevant parameters to help us with predicting future returns. The average actual historical return for bonds is shown by the black, dashed line below.

Before I get arrested for committing more statistical crimes, if you were to retire today, expecting anything more than 3% in annual returns from a Treasury bond portfolio is outright lunacy. And hence even if the average return for bonds in the past was 5%, we’ll use a static 3% for the bond component of our portfolios during retirement. And while we are at it, we’ll also apply a 30% haircut to the average predicted return for stocks during retirement while preserving the same volatility estimate as in the past.

Why do that? For bonds, it’s clear. Just look at where interest rates are today.

What we can reasonably expect out in the future is either rates remaining the same or rising.

And we know what a rising rate environment does to the price of bonds and hence the 3% total return assumption for the bond component of our portfolios.

Paying interest on bonds (issuing debt to finance operations) is a cost to businesses and a rising rate environment means that the cost to service that debt will rise as well. That implies a decline in profitability for businesses that rely on debt financing and that along with where the stock market valuations are today means a 30% haircut on future stock market returns assumption is quite reasonable.

Using these corrected return estimates and past volatility measures, we predict what future stock market returns could look like.

That intermittently random pattern of returns is what we could typically expect though this is just one sample. But what if the sequence of stock market returns of the future follow this pattern?

Or this?

Shown below is what a $1.25 million portfolio invested in a 60/40 stock/bond mix that is re-balanced annually grows to during 35 years in retirement for the three patterns of stock market returns described above. Remember that the bond component is assumed to yield a static 3% during this entire time-frame.

Why start out with a $1.25 million portfolio? That’s because this number assumes a $50,000 inflation-adjusted income draw for each year in retirement and the so-called 4% rule for withdrawal rate at the start gets us to a portfolio size of $1.25 million.

So regardless of the sequence of returns, the final value of the portfolio is the same. And that’s because we are not drawing income from this portfolio yet and hence is left to compound for all those years in retirement.

But this is what happens if we were to draw a 4% annual inflation-adjusted income stream off of the starting portfolio balance.

If the stock markets crater first like what would have happened to us if we were unlucky enough to retire say in 2007 or any other prior stock market peaks, we could run out of money very quickly. And that’s with doing everything right. It’s just that we were dealt a bad hand of the returns distribution.

So what do we do? We save more where instead of relying on say drawing 4% from our portfolios, we get by on drawing 3%. Or even 2%. Why? Say instead of a $1.25 million portfolio to start with, if we had saved up double that amount (ouch), that same $50,000 in income need is a 2% inflation-adjusted withdrawal rate. And that could be had from the dividends and interest payments alone without the need to touch principal. Heck, if that is our income need on a $2.5 million portfolio, we can skip owning bonds in entirety and just live on stock dividends in perpetuity and still have plenty left (if that is our goal).

If doubling of savings is not a possibility, another option as highlighted by Dr. Wade Pfau, Professor of Retirement Income Planning at The American College of Financial Services in this piece is to use a rising glidepath approach to stock allocation while simultaneously reducing the bond component of our portfolios in retirement. That’s counter to what traditional asset allocation models recommend but what this strategy entails is starting out with a very low allocation to stocks right when we retire and gradually increasing that to say 100% stocks towards the later stages of our life in retirement. That’s not likely to completely eliminate the risk of running out of money but will greatly improve the odds of being able to sustain our lifestyles during the entirety of our retirement, so the paper says.

Here’s an example of what happens with the three portfolio return scenarios when we start out with a 10% allocation to stocks and incrementally increase that to 100% stocks through retirement.

So now, instead of running out of money in say year 10 for the worst-case sequence of returns pattern, we were able to extend our income drawing time-frame by double the number of years.

But this apparent safety does not come free as seen by the ending portfolio values for the other two scenarios.

So that was a lot of number crunching and pretty plotting but in an environment where future capital market returns are expected to be low, saving more buys us that ticket to not becoming a victim to an outlier event. Because to quote from that same book by Peter L. Bernstein again,

The essence of risk management lies in maximizing the areas where we have some control over the outcome while minimizing the areas where we have absolutely no control over the outcome and the linkage between effect and cause is hidden from us.

So market returns will be whatever they will be but we know this one thing that still remains in our control and that is how much we save. And of course, the sooner we start, the more time we have for the money to compound and the easier the going gets.

Thank you for reading.

Until later.

Cover image credit – Artem Bali, Pexels

So that’s for the U.S. Then you do the same for the international stock markets and then the bond markets to some extent and so on.

But in this day and age where a company or an industry can be disrupted out of existence in time-frames that are much shorter than what’s typically required to achieve your goals, that old way of approaching portfolio construction needs a bit of refinement.

But before that, some general strategies on how I think about portfolios. I follow a core and explore approach to portfolio construction where the vast majority of my family savings forms the core with the remaining and a very tiny portion at the margins forming the explore. The core is what’s going to take care of my family’s income needs when we need it. The explore will either knock the ball out of the park or will strike out. Don’t care. But for you, unless you have a solid core, there is no reason to do explore. I do it because I am at a point where the absence of explore is not going to change anything in my family’s life. This explore also fulfills some of the alternative investment bucket option within my overall family’s financial plan.

What’s in explore will trickle in here and there from time to time but the core is what I will address here. The global market portfolio excluding hard assets is about 50% stocks and 50% bonds. That is what we all own collectively on this planet. So that becomes our baseline. Now depending upon where you are in life, you might have to adjust that stock-bond mix to get to a ratio that is just right for you at a given time. Setting aside discussion on bonds for now, the stock component of the global market portfolio as it stands today is split almost evenly between U.S. and international (ex-U.S.). Now since I live in U.S., spend my money in U.S. and will likely retire in U.S., I need to design in a U.S. tilt to my portfolio. Some call it a home country bias though tilt is a much milder form of home country bias but that again depends upon where I am going to spend my money…eventually. But once I have the U.S. vs. international proportions defined, I can start getting into the nuts and bolts of what to populate my portfolio with.

To do that, from time to time, I compile datasets of portfolio holdings from a variety of investment providers that allows me to dig deep to assess what each of them have to offer. So I use a lot of copy / paste in bringing that data into Excel but you can use any tool of your choice that hopefully allows you to automate this drudgery.

So back to this, a few thoughts on populating the U.S.-centric component of my portfolio. The easiest way to get that exposure is to buy the entire U.S. stock market in one fell swoop. I like the index compiled by the Center for Research in Security Prices (CRSP) for its completeness and which Vanguard has adopted so I’ll use that dataset to point out a few things.

• You buy this and you buy stakes in 3,618 U.S.-based companies as of this writing.
• The top 10 companies use up 17.8% of your money in this investment.
• The top 50 companies occupy 42% of your money and the top 100 companies, 55%.
• The bottom 3,518 companies take the remaining 45%.

The other popular investment that most of us are familiar with is the S&P 500 index. You buy this and you get to own 500 of the biggest companies in the U.S. because it is after all a large-cap index. Cap by the way is short for market capitalization, the market value of a company.

But S&P 500 faces the same set of issues as the total market index described before does where the ownership is very top-heavy with only a small fraction of your dollars going to mid and small-size companies.

So what’s wrong with a top-heavy investment portfolio? If you believe in pure meritocracy, nothing. The top companies got to where they are because they evolved to dominate their industries, crushing their competition along the way. Or they created something totally new that changed the way we live and the market rewarded them for that.

But there probably was a natural progression for most of these companies from small upstarts to becoming mid-size and then becoming the top dog. So if you buy only this one investment, you miss that transition and the associated growth from small to mid and from mid to a large company.

Plus being so top heavy has its own set of risks as any dislocation in one or two of the companies at the top can have an outsize impact on your portfolio. And I have a personal bias against how some of these top companies are run because I’ve dealt with many of them in one setting or the other. They become too big, too lethargic, too bureaucratic and are an easy prey to nimble upstarts. So capitalism and its forces of creative destruction will replace these companies that are at the top today with companies of tomorrow and you want to participate in that process.

So then you start to look beyond the investments that are large-cap heavy to get exposure to mid-size and smaller companies for an ‘optimized’ portfolio blend.

To get to that, I go back to using the CRSP indices again that target these market segments. There is a third type of investment that I use to get exposure to small/mid-size companies through an extended market or the completion index. The extended market portfolio comprises of all the companies that are found in the total stock market index but without the companies in S&P 500.

A few things I found that I feel you’ve got to know if you care to know .

• All companies found in S&P 500 are also found in the total market index. That’s expected so no big surprise there. But when you buy only total market index, 83% of your money goes to companies in S&P 500 with only 17% devoted to small and mid-size companies.
• Companies in the extended market index and the S&P 500 index are almost 100% mutually exclusive. Knew that but wanted to verify. So if you own one index, you completely exclude the companies from the other index. If you own CRSP total market index, you own both though not in the same proportion so your exposure to small and mid-size companies is limited.
• The overlap between mid-cap and small-cap CRSP portfolios is negligible with only 26 companies that are common to both indices. 3% of your money in a mid-cap portfolio overlaps with holdings in small-cap whereas 2.7% of your money in small-cap overlaps with holdings in mid-cap. So both of these are almost mutually exclusive.
• Out of the 366 holdings in the mid-cap index, 256 of them are already present in S&P 500 so that’s a pretty big overlap. In % terms though, if you own S&P 500, only 14.5% of your money is invested in companies that belong to the mid-cap index. But if you own the mid-cap index, 78% of your money goes to companies that are also found in the S&P 500 index though the proportion of your dollars to mid-size companies is quite a bit larger. So if you want to balance out your portfolio exposure to more nimble yet financially stable companies, you’ve got to have exposure to mid-cap companies. This helps spread out your dollars across a broader set of companies than just the large-caps.
• There are 3,288 companies in the completion index though only 111 could be found in the CRSP mid-cap portfolio. This makes sense as a lot of the companies in the mid-cap portfolio are also found in S&P 500 and none of the S&P 500 companies are found in the completion index. In percentage terms, if you buy the completion index thinking that you are buying exposure to the mid-cap space of the U.S. stock market, you might want to rethink that because only 20% of your money in this investment gets you that mid-cap exposure. On the flip side, if you buy the mid-cap portfolio, only 23% of your dollars are invested in companies that are found in the completion index. Plus oftentimes, you miss exposure to quite a few fast-growing and prominent companies if you decide to buy one or the other.
• So what’s in the completion index then? But before that, a bit about the CRSP small-cap index. There are a total of 1,408 companies in that index out of which 1,373 can also be found in the completion index. So the completion index which is usually thought off as a mid + small-cap index is truly a small-cap index. In % terms, if you own the completion index, 70% of your money goes to companies that are already in the small-cap index so if you own both, you have an outsize exposure to smaller names and not much to mid-size companies. If you own the small-cap index instead, almost 96% of your money goes to companies that are also in the completion index. So when and where if I have access to the completion or the extended market index, I buy that instead of the small-cap portfolio. Plus you get exposure to 1,821 more companies in the completion index that are not found in either the small or mid-cap indices. In % terms, that’s 12.6% of your money in the completion index that goes to companies that are not in either of the two indices.

So that covers the domestic front, on to international…

The easiest way to populate the international bucket is to buy the entire basket of international stocks in one shot through an investment that tracks the FTSE global all cap ex-US index. You buy that and you pretty much own all publicly traded companies in every country outside the United States. This is not as egregiously top-heavy as what you can find with the total U.S. stock market tracker but it has the same type of over-concentration at the top issues. A few stats about this particular investment…

• You buy this and you own stakes in 6,372 companies outside U.S. spanning almost every country around the world.
• The top 10 companies use up 8% of your investment dollars.
• The top 50 companies occupy 22% of your international stock portfolio and the top 100 companies, 32%.
• The bottom 6,272 companies take up the remaining 68%.

To optimize further and to get exposure to potentially more volatile yet faster growing companies, I split this one investment into two parts –

• FTSE all-world ex-U.S. portfolio that provides me exposure to larger, more established companies, 2,761 in total.
• FTSE global small cap ex-U.S. portfolio with exposure to 3,612 small and mid-size companies.

To increase tilt in favor of faster growing emerging markets, I add a third element to this mix of international stocks through an FTSE emerging markets all-cap portfolio.

Here are my findings when I dug deep into the holdings of these four different investment options…

• Pretty much everything in FTSE all-world ex-U.S. and FTSE global small-cap ex-U.S. portfolios is already in FTSE global all-cap ex-US index.
• Only 10% of my money goes to smaller-cap international companies when I buy FTSE global all-cap ex-US index. The remaining 90% of my money then is invested in companies that are found in FTSE all-world ex-U.S. portfolio (its large-cap brethren).
• There is almost no overlap between FTSE all-world ex-U.S. portfolio and FTSE global small cap ex-U.S. portfolio. That helps me to increase exposure to smaller companies if I need to by increasing allocation to the latter.
• Out of the 4,616 companies in FTSE emerging markets all-cap portfolio, 1,115 can be found in FTSE all-world ex-U.S. portfolio and 1,165 in FTSE global small-cap ex-U.S. portfolio. Add the two and about that is what you find in FTSE global all-cap ex-US index from the emerging markets angle. That leaves about 2,336 companies that I don’t have exposure to if I were not to include the emerging market segment in my portfolio.
• About 20% of my dollars are devoted to emerging market stocks in FTSE global all-cap ex-US, FTSE all-world ex-U.S. and FTSE global small cap ex-U.S. portfolios. The remainder in all three is exposed to companies in the developed world outside the United States. That allocation is quite right but I tend to employ a slight tilt in favor of emerging market economies and hence this added option.
• 87% of my money in FTSE emerging markets all-cap portfolio is invested in companies that are also found in FTSE all-world ex-U.S. portfolio, 8.5% of it is invested in companies within FTSE global small cap ex-U.S. portfolio and the remaining 4.5% is invested in companies that are only found in the emerging market segment. That as mentioned before, is invested in the 2,336 companies that are exclusive to the emerging market component of the total portfolio.

So there, some general thoughts on portfolio design. I have intentionally not listed specifics on what I own and in what proportions because every situation is different. Plus these portfolios and proportions evolve over time along with changes in the markets and your goals so its applicability is likely different for everyone. Though I tend to think that this is it, the task is never really, really done because in the words of Salvador Dali, perfection will always remain elusive.

Have no fear of perfection – you’ll never reach it.

But I’ll keep trying.

Until later.

Cover image credit – Bruin, Flickr