Historical Asset Returns & Future Returns

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
Topic Author
SimpleGift
Posts: 3869
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Historical Asset Returns & Future Returns

Post by SimpleGift »

Nearly everything we know about the historical returns of stocks and bonds worldwide has been learned over the past 150 years, thanks to the Dimson, Marsh and Staunton global database (1900-2015), the Ibbotson U.S. data (1926-2015) and the Shiller U.S. database (1871-2015). This post makes the case that all of this asset return data derives from a period in world history when the two primary drivers of global growth — namely, population and labor productivity growth rates — were rapidly increasing. However, since we're now living in an age when both of these primary growth drivers are declining, with no apparent signs of future increase, the question arises: Just how relevant can historical stock and bond returns be for future returns?


1. Between 1900 and 1970, the World's Population Grew Much Faster than Previously in History
World history can be divided into three periods with distinct population growth trends. The first period (pre-modernity) was a very long age of very slow population growth. The second period, beginning with the onset of modernity around 1900 (with rising living standards and improving health) and lasting until about 1970, saw a rapidly increasing rate of growth. Now that period is over, and we are in a third phase: the global population growth rate is falling and will continue to fall, leading to an end of growth before the end of this century.
  • Annual Growth Rate of World Population (% Change), 1800-2025
    Image
    Source: Wikipedia
2. Between 1900 and 1970, the World Had a Great Wave of Productivity Growth That's Now Declining
From 1900 to the First World War, productivity grew moderately. The U.S. then led a major wave of productivity growth in the 1930s and 1940s, while other countries struggled in the aftermath of the Great Depression. Following the Second World War, Europe and Japan enjoyed a golden age, catching the wave that was launched in the U.S. Despite a temporary rise in global productivity in the 1990s from the tech revolution, the trend is now downward.
  • Annual Growth Rate of Labor Productivity (% Change), 1890-2012
    Image
    Source: OECD Observer
IN SUM: There's been much discussion in the financial press (and in several recent Boglehead threads) about the future "normalization" of interest rates and about the "historical" equity risk premium. My concern is that everything we know about "normal" asset returns comes from a period that was apparently quite abnormal in human history — a period which certainly no longer exists and which seems unlikely to recur anytime soon. Your thoughts?
Last edited by SimpleGift on Fri Aug 21, 2015 1:16 pm, edited 2 times in total.
Clive
Posts: 1950
Joined: Sat Jun 13, 2009 5:49 am

Re: Historical Asset Returns & Future Returns

Post by Clive »

Japan might provide a forerunner indication http://video.uk.businessinsider.com/1fe ... 4d722.webm
User avatar
Topic Author
SimpleGift
Posts: 3869
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Historical Asset Returns & Future Returns

Post by SimpleGift »

In a PM, someone asked about emerging markets and their population and productivity growth rates — and whether these might return the world to "normal" economic growth and asset returns in the near future.

However, population growth rates in emerging economies, in aggregate, also peaked around 1970 (in orange, chart below). Granted, these declining growth rates are within very large populations, so the total number of people in emerging economies is still expanding significantly — as is their relative wealth — but their rate of population growth continues to fall.
Productivity growth rates in emerging countries (in orange below) did increase rapidly between 1985 and 2007, but have been on a downward trend ever since. According to The Conference Board, this is the result of the gradual shift from manufacturing to service industries, plus lower overall business confidence after the financial crisis.
In short, given current trends, it doesn't appear that the emerging economies will be returning the world to the pre-1970s days of increasing population and productivity growth rates — or historical asset returns.
User avatar
Topic Author
SimpleGift
Posts: 3869
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Historical Asset Returns & Future Returns

Post by SimpleGift »

In researching this topic, I ran across a related and fascinating insight from the McKinsey Global Institute that I thought was worth adding to this thread (though it's been posted before).

Since the global population and workforce growth rate is now steadily decreasing (an inexorable reality at this point), to drive future increases in global GDP, the entire onus is upon ever expanding productivity growth (chart below). Even if productivity growth stays at its long term average of 1.8%, future GDP growth will still be declining. But global productivity has also been falling for decades, with no ready signs of increases in sight. The apparent result: slower future global economic growth, lower interest rates and a more challenging environment for equities.
BOTTOM LINE: Implicit in every analysis of historical asset returns (a favorite pastime of Bogleheads, including myself) is that the future will resemble the past — i.e., that the stock and bond returns of the 20th century (when all the data was gathered) will have relevance for the 21st century (when we are investing our money). I'm less and less certain this assumption is true.
needtosave
Posts: 16
Joined: Wed Aug 19, 2015 10:31 pm

Re: Historical Asset Returns & Future Returns

Post by needtosave »

fascinating post
User avatar
tarheel
Posts: 412
Joined: Fri Sep 17, 2010 8:33 am

Re: Historical Asset Returns & Future Returns

Post by tarheel »

Nice job as always Simplegift. If us young people needed any more motivation to save as much as we can, here it is.
IlliniDave
Posts: 2328
Joined: Fri May 17, 2013 7:09 am

Re: Historical Asset Returns & Future Returns

Post by IlliniDave »

Well, for someone with a very long investing horizon, maybe a return to the emphasis on "stocks for dividends" is the ultimate destination.
Don't do something. Just stand there!
Mureke
Posts: 36
Joined: Mon Jun 09, 2014 7:50 am

Re: Historical Asset Returns & Future Returns

Post by Mureke »

Somewhat off topic, but I'd like to point out that when using statistics rather than expert opinion, it is not indisputable that world population will stabilize just yet: http://www.sciencemag.org/content/346/6206/234

The main culprit is Africa where population growth rates have not been declining as anticipated. Although world population is not growing as fast as before, it might level off only later than 2100. Of course, overpopulation could cause environmental and political problems that would hamper productivity instead of increasing it.
  • World population forecasts and probability
    Image
    Source: Sciencemag
Call_Me_Op
Posts: 8002
Joined: Mon Sep 07, 2009 2:57 pm
Location: Milky Way

Re: Historical Asset Returns & Future Returns

Post by Call_Me_Op »

While I agree that we cannot rely on past data to predict future returns, the relative performance of asset class categories is likely to hold in the future - that is, small value is likely to return more than large growth over long time periods.
Best regards, -Op | | "In the middle of difficulty lies opportunity." Einstein
protagonist
Posts: 6762
Joined: Sun Dec 26, 2010 12:47 pm

Re: Historical Asset Returns & Future Returns

Post by protagonist »

Simplegift wrote: My concern is that everything we know about "normal" asset returns comes from a period that was apparently quite abnormal in human history ... Your thoughts?
Exactly. The past 150-200 years or so (post-industrial revolution) may well have been the most remarkable period of growth in the history of civilization. And history is cyclical, not linear. Think about the guy in the year 200, when Rome was near its peak after c. 100 years of Pax Romana, who took advice from Marcus Aurelius and sunk all his money into the "All-Empire Stock Index Fund". It might have taken the next 1650 years or so before he broke even.

People who look at past data to make assumptions about the market are usually looking, at best, at 90 years or so of reliable data and, based on that, are drawing assumptions about the next 30. The analogy I have used in the past on this forum is that it is like not taking an umbrella with you Thursday because it did not rain Monday through Wednesday.

I'm not at all a pessimist about the market or I wouldn't have half my assets invested in it. But nor am I an optimist. I realize that we are dealing with a chaotic, complex system and we are extraordinarily fortunate to be living in the most powerful nation on earth during the most extraordinary growth period in history. Sooner or later all things must come to an end. It could happen tomorrow or it could happen a thousand years from now. And either way, there is no even vaguely reliable statistical method that will give you a clue about where the market will be 20 or 30 or 50 years from now...we are just bozos on this bus. I invest the way I do because, basically, I have no clue about the future. And if you have no crystal ball, you'd might as well take the path of least resistance, guess, and hope for the best.

What I have to say to those here who think there is any statistical validity to financial modeling a complex, nonlinear system:
"There are more things in heaven and earth, Horatio, than are dreamt of in your philosophy".

To which I add, all the world is but a play. Be thou the joyful player. (sic)
User avatar
Topic Author
SimpleGift
Posts: 3869
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Historical Asset Returns & Future Returns

Post by SimpleGift »

tarheel wrote:If us young people needed any more motivation to save as much as we can, here it is.
Yes — however the wildcard in all this is how closely future global asset returns will be tied to future GDP growth. With bonds, it's easier to make the case that higher economic growth leads to higher real interest rates, since investment (the demand for capital) tends to be higher than savings (the supply of capital) — the exact opposite of what we have today.

For equities, the connection between higher economic growth and higher stock returns is not quite so direct, but it still holds on an aggregate global level, I believe. In a recent Yearbook, Dimson, Marsh and Staunton concluded: "Many investors and commentators have misunderstood the evidence on economic growth and equity performance. Though difficult for investors to capture in portfolio returns, stronger GDP growth is generally good for investors."

So on a global scale, over the long term, I believe we can say that good economic growth should lead to good assets returns, even though there's not a 1:1 relationship — and that slower economic growth would generally result in lower interest rates and weaker stock returns.
User avatar
Topic Author
SimpleGift
Posts: 3869
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Historical Asset Returns & Future Returns

Post by SimpleGift »

protagonist wrote:
Simplegift wrote: My concern is that everything we know about "normal" asset returns comes from a period that was apparently quite abnormal in human history ... Your thoughts?
Exactly. The past 150-200 years or so (post-industrial revolution) may well have been the most remarkable period of growth in the history of civilization.
Slightly off topic, but what surprised me most in researching this subject was the anemic effects the 1990s tech revolution has apparently had on global productivity growth. According to the charts upthread, U.S. productivity grew from 1% in 1980 to over 2% in 2000 — but has declined ever since, back to just under 1% today. Perhaps the productivity benefits of the 1990s tech revolution are yet to be realized, but so far, they don't appear to have had a substantial impact on global growth.

Contrast this with huge productivity gains of the 1930s, 1940s and 1950s (again, charts upthread), where electricity, the combustion engine and pharmaceuticals boosted productivity by more than 3%-4% in many advanced economies. Unless tech innovation leads to some real breakthroughs in the decades to come (and not just apps and social networking sites!), it's difficult to see how economic growth and asset returns can come close to those of the previous century.
protagonist
Posts: 6762
Joined: Sun Dec 26, 2010 12:47 pm

Re: Historical Asset Returns & Future Returns

Post by protagonist »

Simplegift wrote:
protagonist wrote:
Simplegift wrote: My concern is that everything we know about "normal" asset returns comes from a period that was apparently quite abnormal in human history ... Your thoughts?
Exactly. The past 150-200 years or so (post-industrial revolution) may well have been the most remarkable period of growth in the history of civilization.
Slightly off topic, but what surprised me most in researching this subject was the anemic effects the 1990s tech revolution has apparently had on global productivity growth. According to the charts upthread, U.S. productivity grew from 1% in 1980 to over 2% in 2000 — but has declined ever since, back to just under 1% today. Perhaps the productivity benefits of the 1990s tech revolution are yet to be realized, but so far, they don't appear to have had a substantial impact on global growth.

Contrast this with huge productivity gains of the 1930s, 1940s and 1950s (again, charts upthread), where electricity, the combustion engine and pharmaceuticals boosted productivity by more than 3%-4% in many advanced economies. Unless tech innovation leads to some real breakthroughs in the decades to come (and not just apps and social networking sites!), it's difficult to see how economic growth and asset returns can come close to those of the previous century.
You are pessimistic about the future. You can certainly come up with many good reasons.

I am not. I am realistic about the future. I'm way curious, but I can say without hesitation I have no clue. 2100 could be a post-Armageddon world or it could be like something out of Star Trek or the Jetsons.

Sometime in the past I posted a graph that was published by some Scandinavian bank showing that, since the year 1200 or 1300 or something like that, essentially housing has just kept up with inflation. How they came up with something like that is beyond me, and don't ask me to confirm its validity. But I would not be surprised , if you looked at the entire course of human history since the last ice age and the birth of the first cities, no matter HOW one "invested" (including gold, or the hypothetical stock or bond market, or sheep/cattle, or sticking money under your pillow, whatever) , the gross sum over time might be zero. Some things are more volatile than others, and reward is proportional to risk. Stocks are more volatile than cash, and thus more prone to boom or bust. Recently history has favored boom over bust which is why they have averaged 8% or 10% or whatever return over the last paltry hundred years or so.

People are alive who have lived longer than all the reliable data we have on the stock market. Think about that when you rely on "past data" to plan for the next 20 or 30 or 50 years of your life. It is a house of cards.
User avatar
Topic Author
SimpleGift
Posts: 3869
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Historical Asset Returns & Future Returns

Post by SimpleGift »

protagonist wrote:But I would not be surprised, if you looked at the entire course of human history since the last ice age and the birth of the first cities...
The critically important difference is that through all of the historical time period that you mention, both population and productivity growth rates have been rapidly increasing (see this thread) — but since about 1970, both of these factors have been in a long-term declining trend, with no ready signs of increase. At least for population growth rates, their long-term global decline is already "baked in" and the only question now is the relative rate of decrease.

Since "economic growth = population growth + labor productivity growth", we are truly entering an age that's not been experienced before in the long arc of human history — and one that is certainly not reflected in the historical record of stock and bond returns.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Historical Asset Returns & Future Returns

Post by Rodc »

So, for the sake of argument, what is the action?

This might put a long term damper on returns. When does that start? Is it reasonable to hold a historical portfolio and expect more or less historical behavior for the next 10 years? 20 years? Do I simply adjust my return expectations down and simply save a little more (how much more?)? If I am retired do I reduce withdrawals (to what and how does my expected lifespan play into this?)? Do I shift to some non-standard allocation and how would I know what that is?

The world is certainly changing. Has been for quite some time. Help my understand what actions I should take and why.

Thank you.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
User avatar
Topic Author
SimpleGift
Posts: 3869
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Historical Asset Returns & Future Returns

Post by SimpleGift »

Rodc wrote:So, for the sake of argument, what is the action?
Don't think it can be expressed any better than Garland Whizzer in this post :
garlandwhizzer wrote:I tend to agree … that both economic growth and stock returns are likely to decrease in the next 10 - 50 years relative to the their robust rates in the past century. Having said that, it is clear that the future is both unknown and unknowable. The actionable information here is not to change ones well thought out asset allocation plan, but simply to consider revising future return expectations. Longer lifespans plus lower investment returns translates into working longer, saving and investing more, and spending less both pre- and post-retirement. It may not be news we want to hear, but I suspect it is wise to take this into account in planning for future retirement.
To my mind, it also brings into question many common investing assumptions that are based on the historical asset returns of the past 100 years, such as the "normalization" of future interest rates, the historical equity risk premium and the 4% safe-withdrawal-rate. Not that I've become a pessimist, but I do feel all of these assumptions bear re-examination when looking at the investing climate in the century ahead of us — rather than the century behind us.
User avatar
Maynard F. Speer
Posts: 2139
Joined: Wed Mar 18, 2015 10:31 am

Re: Historical Asset Returns & Future Returns

Post by Maynard F. Speer »

This is why I generally describe myself as a pessimist - and to add to the weight of concern, it worries me that markets have been driven almost entirely by debt, speculation and central banks so far this century ... with little evidence it's done much for the economy

On the other hand, I don't think any of us can anticipate the effects of robotics and artificial intelligence ... Is it too simplistic to see population growth and age as a future constraint on productivity? Rio Tinto has these skyscraper-sized vehicles driving themselves around Australia .. Plus we've got another 5 billion people to accommodate on the planet soon

I may revise all of these, but I think all of these may produce positive returns when broad indexes aren't:
- Take advantage of long-term mean reversion in stock valuations;
- Take advantage of alternative finance (e.g. money lending offers stable returns from 5 to 15%);
- Invest in regions with young/growing populations (India, Turkey, Africa);
- Private equity and active micro-caps - disruptive technologies, high risk/reward, less global exposure;
- Sector bets: global infrastructure and agriculture (we're going to have to feed another 5 billion soon)
- Alpha: hedge funds, relative value, stock-picking, inefficient markets
"Economics is a method rather than a doctrine, an apparatus of the mind, a technique of thinking, which helps its possessor to draw correct conclusions." - John Maynard Keynes
Clive
Posts: 1950
Joined: Sat Jun 13, 2009 5:49 am

Re: Historical Asset Returns & Future Returns

Post by Clive »

The UK's FT250 - next 250 largest stocks by market cap under the FT100 (100 largest stocks) has a recent market cap of around $550Bn (£340Bn) - much much smaller than the FT100. In the scale of things the UK FT250 is comparable to US small cap.

Total market index or even large cap index might more reflect demographics. For smaller cap however much less so. Typically over the mid to longer term a initial equal weighted bunch of stocks will end with just a few of the original still trading as-is, but having made strong (above average) gains. 1984 Shell for instance, a initial FT30 index component in more recent years had grown its market cap to 30% more than the combined total FT30 market cap back in 1984.

Declining population/total earnings might not adversely affect small cap indexes, where a few contribute the largest proportion of total index gains - lifting the index higher. Such characteristics are as likely to persist into the future as in the past - perhaps even more so given the increasing speed of tech/biotech advances.
protagonist
Posts: 6762
Joined: Sun Dec 26, 2010 12:47 pm

Re: Historical Asset Returns & Future Returns

Post by protagonist »

Simplegift wrote:
protagonist wrote:But I would not be surprised, if you looked at the entire course of human history since the last ice age and the birth of the first cities...
The critically important difference is that through all of the historical time period that you mention, both population and productivity growth rates have been rapidly increasing (see this thread)
This statement, on the surface, seems so incredulous that I had to check the source. Can one really believe, for example, that there was more rapid growth during the decline of the Roman empire and the subsequent dark ages than when the empire ramped up to its peak in the second century?

Note how the historical periods were cherry-picked. In year 1, Rome, under Augustus, was barely ekeing its way out of 150 years of almost continual civil war. The baseline interval in the graph ("0 growth") was years 1-1000. So any phenomenal growth in the first two hundred or so years in Europe was conveniently cancelled out by the decline and dark ages. The next interval represented a huge 500 year period, including, for example, the years of the great plague which killed an estimated "30-60% of Europe's population". Needless to say, what happened in the mid-14th century would have more than averaged out any growth in the preceding three centuries. 1500-1600, the next period, was the period of the Reformation, and much of Europe was involved in religious battles. The religious wars came to a head in the fourth period, 1600-1700, and GDP growth actually declined significantly (the graph cannot mask that).

As industry started to gain steam in the 18th century, and went full throttle in the 19th and 20th ( due to the industrial revolution), GDP growth increased dramatically. But what does the graph show during this period? That growth was by no means linear and had its ups and downs, with the last two peaks being 1950-1974 and 1994-2010.

The last five years, 2010-2015 (interesting how they were not averaged into the 1994-2010 interval....more cherry picking to make a point I assume...) have been anemic. So what? I wonder what growth looked like in the US from, say, 1929-1934, or 1860-1865.

Plus which, you stated....
The critically important difference is that through all of the historical time period that you mention, both population and productivity growth rates have been rapidly increasing
Really? The historical period I mentioned was "since the last ice age". That was at least 11000 years ago. This graph covers no more than 9% of the historical period I mentioned.

The graph says nothing meaningful, other than that the past 200 or so years have been remarkable, which we already know. And that the dark ages were a lousy time to be alive in Europe.

Statistics don't lie. Statisticians do.
Last edited by protagonist on Sun Aug 23, 2015 3:58 pm, edited 1 time in total.
User avatar
Maynard F. Speer
Posts: 2139
Joined: Wed Mar 18, 2015 10:31 am

Re: Historical Asset Returns & Future Returns

Post by Maynard F. Speer »

Clive wrote:The UK's FT250 - next 250 largest stocks by market cap under the FT100 (100 largest stocks) has a recent market cap of around $550Bn (£340Bn) - much much smaller than the FT100. In the scale of things the UK FT250 is comparable to US small cap.

Total market index or even large cap index might more reflect demographics. For smaller cap however much less so. Typically over the mid to longer term a initial equal weighted bunch of stocks will end with just a few of the original still trading as-is, but having made strong (above average) gains. 1984 Shell for instance, a initial FT30 index component in more recent years had grown its market cap to 30% more than the combined total FT30 market cap back in 1984.

Declining population/total earnings might not adversely affect small cap indexes, where a few contribute the largest proportion of total index gains - lifting the index higher. Such characteristics are as likely to persist into the future as in the past - perhaps even more so given the increasing speed of tech/biotech advances.
There's a similar argument here - I think at least you can say innovation at the very small level will tend to be affected by quite different factors and tailwinds ... I don't normally base investing decisions on macroeconomic predictions, but I think there's a good case when you look at fundamentals, and certainly a good case for diversification

Image
"Economics is a method rather than a doctrine, an apparatus of the mind, a technique of thinking, which helps its possessor to draw correct conclusions." - John Maynard Keynes
User avatar
Topic Author
SimpleGift
Posts: 3869
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Historical Asset Returns & Future Returns

Post by SimpleGift »

protagonist wrote:The graph says nothing meaningful, other than that the past 200 or so years have been remarkable, which we already know. And that the dark ages were a lousy time to be alive in Europe.
With respect, this thread isn't about debating economic growth through ancient history, but rather encouraging investors to look ahead at the primary economic growth factors (population and productivity growth) in the decades to come — and perhaps re-examine investing assumptions that are based on the historical record of stock and bond returns.

In that vein, for those interested in more about the prospects for future productivity growth, the McKinsey Global Institute has an interesting white paper, Global Growth: Can Productivity Save the Day in an Aging World?, which is one of the best summaries I've found of current thinking.
User avatar
Riprap
Posts: 509
Joined: Thu Jan 29, 2009 2:08 pm

Re: Historical Asset Returns & Future Returns

Post by Riprap »

I think I will always remember a quote from Gus Sauter from an Oct 2012 interview in Barrons,

"A lot of people assume that market returns are driven by economic growth. But that's not really the case. A better predictor is valuation, or price-to-earnings ratios. Right now, the market is about 13 times earnings. Fourteen or 15 is the historical average. So we've got a little bit of a discount. That provides for a higher return than you might otherwise have expected."

http://www.barrons.com/articles/SB50001 ... 0146687174

I believe Warren Buffett essentially says the same thing.

Here's a link to another talk Sauter gives where he discusses stock market returns and economic growth. The correlation is zero he says.

http://fostergroup.markupfactory.com/wi ... mic-growth

I suspect a lot of Bogleheads pay attention when Gus Sauter has something to say.
User avatar
vitaflo
Posts: 1388
Joined: Sat Sep 03, 2011 3:02 pm

Re: Historical Asset Returns & Future Returns

Post by vitaflo »

protagonist wrote: Sometime in the past I posted a graph that was published by some Scandinavian bank showing that, since the year 1200 or 1300 or something like that, essentially housing has just kept up with inflation. How they came up with something like that is beyond me, and don't ask me to confirm its validity.
You don't need to go back that far, just look at the Case Shiller Index of home prices. Adjusted for inflation, a $100,000 house in 1896 would have been worth $106,000 in 1996. It was only since the late 1990's that housing was an "investment" and prices rose above inflation. No wonder we had a crash, because housing prices rising much more than inflation was never something that happened before.

One may be able to say similar things with regards to population growth, productivity, etc. Frankly I see wages being impacted much more than returns. Automation will become a real issue moving forward for many people's incomes. Forget about returns. Hard to get any when so many will be out of work and not able to invest at all.
Last edited by vitaflo on Sun Aug 23, 2015 4:40 pm, edited 1 time in total.
User avatar
Topic Author
SimpleGift
Posts: 3869
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Historical Asset Returns & Future Returns

Post by SimpleGift »

Riprap wrote:Here's a link to another talk Sauter gives where he discusses stock market returns and economic growth. The correlation is zero he says.
No doubt that valuations are an important determinant of stock returns, especially in short run — and that there hasn't been strong historical correlations between individual country's GDP growth and their stock market returns.

But here we are talking about global equities, in aggregate, over long-term investment horizons. Perhaps you missed the quote upthread by Dimson, Marsh and Staunton from their 2014 Yearbook, where they concluded: "Many investors and commentators have misunderstood the evidence on economic growth and equity performance. Though difficult for investors to capture in portfolio returns, stronger GDP growth is generally good for investors."
Clive
Posts: 1950
Joined: Sat Jun 13, 2009 5:49 am

Re: Historical Asset Returns & Future Returns

Post by Clive »

Maynard F. Speer wrote:I think at least you can say innovation at the very small level will tend to be affected by quite different factors and tailwinds ... I don't normally base investing decisions on macroeconomic predictions, but I think there's a good case when you look at fundamentals, and certainly a good case for diversification
The FT100 has lagged quite significantly. Equal weighted selections have generally paced the better FT250 gains since 2000. Equal weighting is the same as a small 'tilt'. Which implies that the overweighted larger caps have induced a drag factor and indeed their earnings have grown around 30% compared to the FT250's earnings expanding 90%. Large caps that falter can remain relatively large, but require time to heal. Cap weighting is a bet on large momentum, fine if that works, not so good if they falter. Equal weighting is more neutral and if anything adds to rewards as smaller companies can more usually expand faster/quicker than larger companies.

I personally prefer to use the average of all companies earnings growth rates as a benchmark i.e. equal weighted (or a close index version such as the FT250). Tilted indexes such as large cap weighted are less representative of the broader average IMO.

Since 1988 for yearly nominal total gains in GB£ percentage terms the S&P500 has produced a 11% average with 18.8 standard deviation compared to FT250's 13.5% average 23.1 standard deviation. As a crude Sharpe type measure dividing the average by the standard deviation produces 0.58 for both. i.e. comparable risk adjusted rewards. FT250 has yielded a higher overall gain as its a small cap index in terms of US scale.

A UK investor holding two thirds FT250, one sixth S&P500, one sixth Berkshire Hathaway and given that the FT250 earns around half of its earnings from foreign, is a form of large, small with 33% exposure to each of GB£, US$ and rest of world. Over the same period that yielded a 13.95 average with 20 standard deviation.

Keeps costs low i.e. VMID (FT250) 0.1% expense, BRK-B 0% ongoing fee following purchase, VUSA 0.07%. US withholding tax of 15% against 2.9% dividend and 1/6th exposure = 0.0725% overhead relative to total portfolio value, expenses amount to 0.0783%, so in combination expenses and taxes of around 0.15%. Not bad IMO for diversification across 750 stocks comprised of small, large, £, $, other currencies etc. I'm more content to hold something along those lines rather than diving into the AIM market - but that's just my own personal comfort zone.
Clive
Posts: 1950
Joined: Sat Jun 13, 2009 5:49 am

Re: Historical Asset Returns & Future Returns

Post by Clive »

vitaflo wrote:
protagonist wrote: Sometime in the past I posted a graph that was published by some Scandinavian bank showing that, since the year 1200 or 1300 or something like that, essentially housing has just kept up with inflation. How they came up with something like that is beyond me, and don't ask me to confirm its validity.
You don't need to go back that far, just look at the Case Shiller Index of home prices. Adjusted for inflation, a $100,000 house in 1896 would have been worth $106,000 in 1996.
...
That's a bit like saying between 1899 and 1985 share price (index only) barely exceeded inflation. Count the dividends and stocks yielded much better real gains. Count the gross imputed rent from a home and house gains somewhat compare if not exceed stocks, at least in the UK.

If I sold my home invested the proceeds in stocks and rented a similar property next door then stocks would have to rise in price to compare to house price rises, and pay a dividend that covered the gross rental yield to rent the property.

Another similarity is land. If you look at just land prices alone and broadly they'll just pace inflation. Grow and sell crops using that land and the land nets a 'dividend'.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Historical Asset Returns & Future Returns

Post by Rodc »

Clive wrote:
vitaflo wrote:
protagonist wrote: Sometime in the past I posted a graph that was published by some Scandinavian bank showing that, since the year 1200 or 1300 or something like that, essentially housing has just kept up with inflation. How they came up with something like that is beyond me, and don't ask me to confirm its validity.
You don't need to go back that far, just look at the Case Shiller Index of home prices. Adjusted for inflation, a $100,000 house in 1896 would have been worth $106,000 in 1996.
...
That's a bit like saying between 1899 and 1985 share price (index only) barely exceeded inflation. Count the dividends and stocks yielded much better real gains. Count the gross imputed rent from a home and house gains somewhat compare if not exceed stocks, at least in the UK.

If I sold my home invested the proceeds in stocks and rented a similar property next door then stocks would have to rise in price to compare to house price rises, and pay a dividend that covered the gross rental yield to rent the property.

Another similarity is land. If you look at just land prices alone and broadly they'll just pace inflation. Grow and sell crops using that land and the land nets a 'dividend'.
Indeed.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
Clive
Posts: 1950
Joined: Sat Jun 13, 2009 5:49 am

Re: Historical Asset Returns & Future Returns

Post by Clive »

A primary future risk IMO is that of taxation risk. In the UK we're told that open borders with Europe yields benefits the UK economy - yet we have the conflict that taxes have risen, benefits declined and wages been suppressed as lower cost migrants compete for units of work. My generation have the benefit of inflation linked pensions etc that younger generations do not as one example. And that's been under expanding population. If/when the population age and taxpayer numbers decline then presumably taxes will need to rise further in order to maintain/expand public spending. Such effects are common to developed states and those states will have to find increasing ways to relieve investors of their wealth to support spending. Or perhaps we might see a reversal away from globalisation and a return to domestic markets applying protectionism.
protagonist
Posts: 6762
Joined: Sun Dec 26, 2010 12:47 pm

Re: Historical Asset Returns & Future Returns

Post by protagonist »

Simplegift wrote: With respect, this thread isn't about debating economic growth through ancient history, but rather encouraging investors to look ahead at the primary economic growth factors (population and productivity growth) in the decades to come
Right-sorry if I got sidetracked, but I thought you were using that chart to reinforce your point about that so I addressed it. And to this point I would say that the future is completely unknowable beyond a few years because noise overwhelms signal. Including by McKinsey.
— and perhaps re-examine investing assumptions that are based on the historical record of stock and bond returns.
And on this point (as I stated in my initial response to your opening post), I couldn't agree with you more. Unfortunately we have nothing better, which pretty much translates to nothing. Other than our beliefs based on limited experience.
protagonist
Posts: 6762
Joined: Sun Dec 26, 2010 12:47 pm

Re: Historical Asset Returns & Future Returns

Post by protagonist »

Rodc wrote: Another similarity is land. If you look at just land prices alone and broadly they'll just pace inflation. Grow and sell crops using that land and the land nets a 'dividend'.

Assuming the land is priced fairly based on its potential for production, if the yield exceeds the factored-in price, it nets a "dividend". If it fails to perform as expected, its value declines.
Last edited by protagonist on Sun Aug 23, 2015 8:30 pm, edited 1 time in total.
garlandwhizzer
Posts: 3022
Joined: Fri Aug 06, 2010 3:42 pm

Re: Historical Asset Returns & Future Returns

Post by garlandwhizzer »

I totally agree with Simplegift's thoughtful analysis of the forces (declining workforce growth, aging demographics, and lowering productivity growth) which if they continue are very likely to challenge our achieving past historical returns on both stocks and bonds going forward. In simple terms economic output depends on only two factors, the number of people working and their average production per worker. Some have pointed out that stock returns do not correlate well with economic growth on a single country basis. True, but when there is a general persistent worldwide pattern of declining economic growth as Simplegift suggests, it is very likely to impact market returns just as the the sudden severe worldwide collapse in stock prices correlated with an equally dramatic and pervasive economic contraction in the Great Depression. It is hard for me to imagine a world in which there is worldwide lowering of workplace participation rates, stagnant productivity growth, persistent low economic growth, and yet in the face of all this, exuberant stock price appreciation.

Simplegift is not alone in his concerns. Many leading investment firms and analysts, among them GMO, Vanguard, RAFI, AQR, Bill Bernstein, Jack Bogle, and Rick Ferri have all indicated that they do not expect average historical returns in US stock or bond markets going forward. They differ somewhat in estimating how much future returns will fall short, but all agree returns will fall short by some measure. Many expect a traditional 60/40 portfolio will produce no greater than 2% - 3% real returns over the next decade or more, and some are expecting closer to zero.

No one can predict the future, anything can happen. That is the good news. Still it may not be wise in financial and retirement planning to count on historical US bond and stock returns going forward. All US assets are by historical standards richly valued now as a starting point. We have enjoyed consistent tailwinds for much of the last half century in both stocks and bonds, but we are likely to encounter some headwinds up ahead. Personally I do not expect a severe collapse like 1929 or 2007-8, but more likely a long slow multi-year slog upward with considerable volatility in the road from time to time.

Garland Whizzer
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Historical Asset Returns & Future Returns

Post by Rodc »

Many expect a traditional 60/40 portfolio will produce no greater than 2% - 3% real returns over the next decade or more,
I don't thing that is the same issue. That is driven far as I can tell simply by current high P/E10 and/or low current dividends. That is simply based on current conditions. That has nothing to do with multi-decade analyses of how the world might change of the sort simplegift is discussing.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
protagonist
Posts: 6762
Joined: Sun Dec 26, 2010 12:47 pm

Re: Historical Asset Returns & Future Returns

Post by protagonist »

Though many posters here seem to have a sense about future returns , none have demonstrated how accurate past projected returns forward have fared. A little back testing would be interesting.

For example, given projections of, say, 20 randomly selected but well-respected "experts" in 1989, what did the distribution of their projections look like for 1990-2000? For 1990-2015? And how well did they do?

In 1990, how many predicted (and factored in the impact of) the S+L crisis? The tech bubble? 9/11? A subsequent 14 year military engagement in Iraq and Afghanistan? Katrina? The housing bubble? The 2008 crash with near collapse of our banking system and the rapid recovery? Desert Storm? The debt crisis in Europe and the history of the euro? The massive rise of China and decline of Japan? The internet economy, personal computing, mass computer trading and smart phones? These are just some of the major events that shaped the markets over the past 25 years. Without a priori knowledge of these events any sophisticated models of the future in 1990 would have been just guesses. As is true today.

I recall another thread where Dr. Bill Bernstein's most recent projected returns going forward were discussed, and somebody posted prior projections of his in 2002 for 2002-2012. The projections seemed, in retrospect, like little more than guesses, and he did not seem to me to do any better than random (within a reasonable range). As would be expected...this is no negative reflection on Dr. Bernstein (who I think most of us respect). And that was only a prediction for 10 years....the ratio of noise/signal grows exponentially with time.

Regarding the relationship between growth and future returns, this is an interesting article: http://www.economist.com/news/finance-a ... th-paradox
User avatar
Riprap
Posts: 509
Joined: Thu Jan 29, 2009 2:08 pm

Re: Historical Asset Returns & Future Returns

Post by Riprap »

Sometimes I just push back away from my desk mesmerized at the at the intellect expended pondering the unknowable.

...and then I get right back to reading :beer
protagonist
Posts: 6762
Joined: Sun Dec 26, 2010 12:47 pm

Re: Historical Asset Returns & Future Returns

Post by protagonist »

Riprap wrote:Sometimes I just push back away from my desk mesmerized at the at the intellect expended pondering the unknowable.

...and then I get right back to reading :beer
++++1 (haha)
Clive
Posts: 1950
Joined: Sat Jun 13, 2009 5:49 am

Re: Historical Asset Returns & Future Returns

Post by Clive »

Own the haystack and investment gain/growth is the same as the haystack growth rate. Whatever the global market cap growth rate is the gain received.

Smaller bales of hay however have the potential to grow bigger faster. Equal weighting each bale of hay (same as small tilt) and the gain received is proportional to the average percentage growth rate across all bales.
Post Reply