The Deep History of Stock & Bond Returns

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
golfCaddy
Posts: 704
Joined: Wed Jan 10, 2018 10:02 pm

Re: The Deep History of Stock & Bond Returns

Post by golfCaddy » Wed Jun 13, 2018 3:16 pm

SimpleGift wrote:
Wed Jun 13, 2018 11:37 am
golfCaddy wrote:
Wed Jun 13, 2018 11:28 am
Even ignoring the quality issues with the stock returns, this glosses over the assumptions in coming up with a risk-free rate of return.
Well, I have to take the professional researchers at their word for the risk-free rates that they've published in the table above (see RF% in the table). They show the risk-free rates in the 3%-4% range for the 1629-1870 period.

When deciding points of fact, and evaluating the difference between peer-reviewed professional research and anecdotal online comments, I'm usually persuaded by the former.
If you don't want to take my word for it, read the quotes below from the Golez paper. They didn't include UK securities in their estimate of the risk free rate until 1727.
Our estimates of the risk rate are based on returns on Dutch and English government bonds that are available from 1650 onwards. Between 1650 and 1720 we use returns on Dutch (redeemable) annuities. This was the most liquid form of Dutch government debt of the time. Data come from Gelderblom and Jonker (2010). There is a gap in our data between 1720 and 1727.
Information for the years before 1650 and between 1720 and 1727 is missing due to data limitations.

GAAP
Posts: 588
Joined: Fri Apr 08, 2016 12:41 pm

Re: The Deep History of Stock & Bond Returns

Post by GAAP » Wed Jun 13, 2018 3:22 pm

siamond wrote:
Wed Jun 13, 2018 2:14 pm
When I made this statement, I had in mind the finer-grain trajectory of what happened in the past, in term of decades, not centuries. Where bonds had flat periods of time, upward periods of time, downwards period of time, with little rhyme or reason. And I defined consistency as return-to-the-mean, and clearly this didn't happen over decades, nor centuries.
So, what is fundamentally different between equity and bonds such that a return-to-the-mean can occur with equities and not bonds?

How can this be true even across changes in monetary systems and standards?

User avatar
knpstr
Posts: 2154
Joined: Thu Nov 20, 2014 8:57 pm
Location: Michigan

Re: The Deep History of Stock & Bond Returns

Post by knpstr » Wed Jun 13, 2018 3:23 pm

SimpleGift wrote:
Wed Jun 13, 2018 2:36 pm
JoMoney wrote:
Wed Jun 13, 2018 2:08 pm
jj45 wrote:
Wed Jun 13, 2018 2:00 pm
...
It's interesting that inflation is so low. When and why did modern-era inflation set in?
When money/bonds stopped being backed by a fixed physical commodity gold/silver standard.
To support JoMoney's point, an historical chart showing the deep history of inflation going back to the 1200s (at left below). Modern Inflation as we know it started in the first half of the 20th century, as countries around the world abandoned commodity-based money (e.g., gold and silver).
Didn't Keynes popularity also largely effect this? His paper "The General Theory of Employment, Interest and Money" was written 1930s which gave rise to how we still deal with the economy today. He encouraged government intervention to combat recessions, namely increased government borrowing/spending even being a proponent of budget deficits for governments. All of which can drive inflation.

Of course being off the gold standard and having a central bank allowed governments to implement Keynes ideas.
Very little is needed to make a happy life; it is all within yourself, in your way of thinking. -Marcus Aurelius

User avatar
siamond
Posts: 4185
Joined: Mon May 28, 2012 5:50 am

Re: The Deep History of Stock & Bond Returns

Post by siamond » Wed Jun 13, 2018 4:02 pm

GAAP wrote:
Wed Jun 13, 2018 3:22 pm
siamond wrote:
Wed Jun 13, 2018 2:14 pm
When I made this statement, I had in mind the finer-grain trajectory of what happened in the past, in term of decades, not centuries. Where bonds had flat periods of time, upward periods of time, downwards period of time, with little rhyme or reason. And I defined consistency as return-to-the-mean, and clearly this didn't happen [for bonds] over decades, nor centuries.
So, what is fundamentally different between equity and bonds such that a return-to-the-mean can occur with equities and not bonds?

How can this be true even across changes in monetary systems and standards?
I have no idea. It was just an empirical observation. I agree it is quite baffling. I guess something fundamental about human nature and their willingness to pay for short-term risk in a reasonably free market... While bonds are much more constrained by the whim of policy makers... Further comments and views welcome!

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: The Deep History of Stock & Bond Returns

Post by SimpleGift » Wed Jun 13, 2018 4:06 pm

golfCaddy wrote:
Wed Jun 13, 2018 3:16 pm
If you don't want to take my word for it, read the quotes below from the Golez paper. They didn't include UK securities in their estimate of the risk free rate until 1727.
Not sure what you're getting at here, but as a cross-check, we can refer to another recent research paper, Eight Centuries of the Risk Free Rate, which reports interest rates from Amsterdam for the 1600-1702 period, and then British console data from the Bank of England for 1703-1919.

Just eyeballing off their chart (p. 15), the average real risk-free rate was about 4.5% in the 1600s, about 4.0% in the 1700s and 3.5% in the 1800s (with considerable short-term variability) — all very much in line with the Golez and other data.
Cordially, Todd

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: The Deep History of Stock & Bond Returns

Post by SimpleGift » Wed Jun 13, 2018 6:56 pm

GAAP wrote:
Wed Jun 13, 2018 3:22 pm
So, what is fundamentally different between equity and bonds such that a return-to-the-mean can occur with equities and not bonds?

How can this be true even across changes in monetary systems and standards?
To take a quick stab at your question: In recent centuries, for countries not devastated by world war, the formula for consistent long-term stock returns has been:
  • Steady GDP Growth --> Steady Corporate Earnings --> Steady Stock Returns

    Image
    Data source: Maddison
So for the 21st century, in the broadest possible terms, as long as global GDP growth continues, we should expect stock returns to follow pace (setting aside issues of share dilution, reinvested dividends, etc.).

For real interest rates on the other hand, on the scale of centuries, they simply reflect the overall risk inherent in the society at the time. It's just not as risky to lend money today as it was in the late Middle Ages. Ironically in a way, the steady economic growth of recent centuries, which has made our world more prosperous, less risky and relatively stable, has reduced the real interest rates on our bond investments!
Cordially, Todd

User avatar
siamond
Posts: 4185
Joined: Mon May 28, 2012 5:50 am

Re: The Deep History of Stock & Bond Returns

Post by siamond » Wed Jun 13, 2018 10:02 pm

SimpleGift wrote:
Wed Jun 13, 2018 6:56 pm
Steady GDP Growth --> Steady Corporate Earnings --> Steady Stock Returns
Well. If everything is (was) steady, there is (was) return-to-the-mean, because this is the definition of steady. This is a circular argument, I am afraid.
SimpleGift wrote:
Wed Jun 13, 2018 6:56 pm
For real interest rates on the other hand, on the scale of centuries, they simply reflect the overall risk inherent in the society at the time.
Yes, this 'societal risk' explanation probably applies at the scale of centuries, but I don't think it gives much of a clue about the erratic trajectory of (real) bond returns over decades. Here is a growth chart I plagiarized from ERN, which I find very striking. Note the green/yellow/red arrows illustrating the point. The trajectory of stock returns, for all its short-term drama, was actually much more consistent in the long-term (over decades - and even centuries, as you illustrated).

Image

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: The Deep History of Stock & Bond Returns

Post by SimpleGift » Wed Jun 13, 2018 11:51 pm

siamond wrote:
Wed Jun 13, 2018 10:02 pm
SimpleGift wrote:
Wed Jun 13, 2018 6:56 pm
Steady GDP Growth --> Steady Corporate Earnings --> Steady Stock Returns
Well. If everything is (was) steady, there is (was) return-to-the-mean, because this is the definition of steady. This is a circular argument, I am afraid.
Was hoping to just present empirical evidence (not an argument), via the colored GDP chart upthread, that GDP growth has been steady as rock in recent centuries, for those countries not damaged by world war. Don't know if there is a straighter line in financial history than the growth of the U.K. economy since 1830 (1.5% CAGR, in blue upthread), or the U.S. economy since 1870 (1.9% CAGR, in green upthread).

As you suggest, this steady GDP growth would support reversion-to-the-mean in stock returns — but not real interest rates.
Last edited by SimpleGift on Thu Jun 14, 2018 12:09 am, edited 1 time in total.
Cordially, Todd

Valuethinker
Posts: 35968
Joined: Fri May 11, 2007 11:07 am

Re: The Deep History of Stock & Bond Returns

Post by Valuethinker » Thu Jun 14, 2018 3:07 am

SimpleGift wrote:
Tue Jun 12, 2018 12:15 pm
Call_Me_Op wrote:
Tue Jun 12, 2018 11:07 am
The obvious take away is not to assume more than 2% real for bonds - which you should not be assuming anyway.
Yes, my sense of the lessons for the 21st century from the deep history data is that 1) it seems unlikely that global stocks would massively underperform 5%-6% real over the next 100 years, and 2) it seems very unlikely that global bond returns would rise back to, say, 4% real in this century, barring a world catastrophe.
The 2 conflict.

If real returns on bonds are low, so too will be real returns on equities? Otherwise you would get massive arbitrage between the 2 asset classes - borrow in the bond markets, invest in the equity markets. That may be precisely what has been happening now via hedge funds or corporate share buybacks.

A caveat is that a period of high inflation could wipe out (nominal) bond returns, but not necessarily equities. That said, in the period of high inflation, equity returns were pretty lousy.
Last edited by Valuethinker on Thu Jun 14, 2018 7:27 am, edited 1 time in total.

Valuethinker
Posts: 35968
Joined: Fri May 11, 2007 11:07 am

Re: The Deep History of Stock & Bond Returns

Post by Valuethinker » Thu Jun 14, 2018 4:38 am

SimpleGift wrote:
Wed Jun 13, 2018 11:51 pm
siamond wrote:
Wed Jun 13, 2018 10:02 pm
SimpleGift wrote:
Wed Jun 13, 2018 6:56 pm
Steady GDP Growth --> Steady Corporate Earnings --> Steady Stock Returns
Well. If everything is (was) steady, there is (was) return-to-the-mean, because this is the definition of steady. This is a circular argument, I am afraid.
Was hoping to just present empirical evidence (not an argument), via the colored GDP chart upthread, that GDP growth has been steady as rock in recent centuries, for those countries not damaged by world war. Don't know if there is a straighter line in financial history than the growth of the U.K. economy since 1830 (1.5% CAGR, in blue upthread), or the U.S. economy since 1870 (1.9% CAGR, in green upthread).

As you suggest, this steady GDP growth would support reversion-to-the-mean in stock returns — but not real interest rates.
I'd have to understand why those endpoints? 1830 & 1870?

Is that simply a function of the data available?

User avatar
legio XX
Posts: 443
Joined: Tue Jul 15, 2008 6:37 am
Location: NYC

Re: The Deep History of Stock & Bond Returns

Post by legio XX » Thu Jun 14, 2018 5:46 am

SimpleGift wrote:
Tue Jun 12, 2018 9:12 am
Just to add more flavor to the deep history of stocks and bonds, beyond the tables and charts:

Early Bond Trading Era — 14th century bankers in an Italian counting house.
This is a depiction of the vice of avaritia (greed). :twisted:
https://www.bl.uk/catalogues/illuminate ... iginID=503

Will have to look at those articles and see how they got their data for the earlier centuries. Legal stuff has an amazingly long half-life.

User avatar
AtlasShrugged?
Posts: 588
Joined: Wed Jul 15, 2015 6:08 pm

Re: The Deep History of Stock & Bond Returns

Post by AtlasShrugged? » Thu Jun 14, 2018 6:27 am

This suggest that, data quality issues aside, investing in the broad universe of available global equities has provided fairly consistent risk-adjusted returns for over four centuries.
SimpleGift....I have greatly enjoyed your history posts; I am an amateur student (dilettante?) of history myself.

On the actionability front, I am wondering if that 50/50 stock to bond guidance should be modified. I am looking at that equity and bond graph and wonder if it is really a good idea to have 50% bonds. My IPS has me on a glidepath to a 50/50 allocation (over the next 15 years). No, I don't plan on living for centuries, but then again, I do not see real bond returns 'moving up' in the last 400 to 600 years. Seems pretty definitive to me.

If 50/50 is not the right allocation going forward, how would we go about determining what would be a better allocation, in your opinion [would love to hear siamond chime in on this question as well]? Looking at the graph, it looks like ~200 years of decline, then leveling off, and another decline. If I am reading the graph right, it looks like we are about to level off for a long time at a very low rate of real return for bonds. That has some implications for retirement planning and bequest motives, no?
“If you don't know, the thing to do is not to get scared, but to learn.”

Valuethinker
Posts: 35968
Joined: Fri May 11, 2007 11:07 am

Re: The Deep History of Stock & Bond Returns

Post by Valuethinker » Thu Jun 14, 2018 7:26 am

AtlasShrugged? wrote:
Thu Jun 14, 2018 6:27 am
This suggest that, data quality issues aside, investing in the broad universe of available global equities has provided fairly consistent risk-adjusted returns for over four centuries.
SimpleGift....I have greatly enjoyed your history posts; I am an amateur student (dilettante?) of history myself.

On the actionability front, I am wondering if that 50/50 stock to bond guidance should be modified. I am looking at that equity and bond graph and wonder if it is really a good idea to have 50% bonds. My IPS has me on a glidepath to a 50/50 allocation (over the next 15 years). No, I don't plan on living for centuries, but then again, I do not see real bond returns 'moving up' in the last 400 to 600 years. Seems pretty definitive to me.
Be a bit more short termist. 15 years is not a long time. Stock markets can easily underperform bonds for 15 years.

Markets have had a 10 year run, almost (not quite) unprecedented in length and performance. Granted, after a smash unprecedented since the 1930s (although bear markets in the 70s actually dropped more, at least adjusted for inflation; the UK market dropped over -80% 1972-74).

The balance of probabilities suggests we are closer to the next bear market than not. It might be your plain vanilla -30% bear, or it might be something grander.

In either case, will you be sanguine at say 60% or 70% equities?
If 50/50 is not the right allocation going forward, how would we go about determining what would be a better allocation, in your opinion [would love to hear siamond chime in on this question as well]? Looking at the graph, it looks like ~200 years of decline, then leveling off, and another decline. If I am reading the graph right, it looks like we are about to level off for a long time at a very low rate of real return for bonds. That has some implications for retirement planning and bequest motives, no?
The short answer is I do not think you can.

I wouldn't base my investment strategy on what the shares of the East India Company or the VOC (Dutch E India co) did in the 1700s. Or what UK gilts paid starting in the War of Spanish Succession (without checking ran 1703-1714).

User avatar
siamond
Posts: 4185
Joined: Mon May 28, 2012 5:50 am

Re: The Deep History of Stock & Bond Returns

Post by siamond » Thu Jun 14, 2018 7:32 am

AtlasShrugged? wrote:
Thu Jun 14, 2018 6:27 am
On the actionability front, I am wondering if that 50/50 stock to bond guidance should be modified. I am looking at that equity and bond graph and wonder if it is really a good idea to have 50% bonds. My IPS has me on a glidepath to a 50/50 allocation (over the next 15 years). No, I don't plan on living for centuries, but then again, I do not see real bond returns 'moving up' in the last 400 to 600 years. Seems pretty definitive to me.

If 50/50 is not the right allocation going forward, how would we go about determining what would be a better allocation, in your opinion [would love to hear siamond chime in on this question as well]? Looking at the graph, it looks like ~200 years of decline, then leveling off, and another decline. If I am reading the graph right, it looks like we are about to level off for a long time at a very low rate of real return for bonds. That has some implications for retirement planning and bequest motives, no?
Well, personally, I don't think 50/50 was particularly wise in the past, nor should it be in the present, for accumulators or retirees. My point has little to do with Simplegift 'deep history' fascinating graph though, and is two-fold:
1. Asset Allocation is a very personal decision. This strongly depends on your personal financial situation, on your primary financial goals, on your primary financial fears, and on your psychology/behavior. It would be foolish to push for a one-size-fits-all in this respect, 50/50 or whatever else.
2. A point I was trying to make in the past few posts is that although there seems to be a strong downwards trend for bonds returns in the 'deep history' (over centuries), the finer-grain trajectory of bonds (real) returns has been remarkably erratic in the 'modern history' (e.g. since 1870, over decades). I have absolutely no clue where bonds returns are going 20 or 30 years from now. While I have more of an inkling about stock returns. In other words, knowledge of the 'deep history' about bonds doesn't seem terribly actionable.

Then there are other considerations on how to properly define and then address meaningful risk factors, that should not speak for a bonds-heavy allocation in retirement imho, but that would be side-tracking the OP's thread. And... again, this is a very personal decision.

PS. I love the 'thing to do' quote you have in your signature. Very meaningful for the topic at stakes!

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: The Deep History of Stock & Bond Returns

Post by SimpleGift » Thu Jun 14, 2018 8:46 am

Valuethinker wrote:
Thu Jun 14, 2018 4:38 am
I'd have to understand why those endpoints? 1830 & 1870?

Is that simply a function of the data available?
Yes. Though there is annual GDP-per-capita data for the United Kingdom going back to 1700 (found here), for most other major economies all the comparable data starts somewhere in the 1800s, as best I know.
Cordially, Todd

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: The Deep History of Stock & Bond Returns

Post by SimpleGift » Thu Jun 14, 2018 10:19 am

AtlasShrugged? wrote:
Thu Jun 14, 2018 6:27 am
If I am reading the graph right, it looks like we are about to level off for a long time at a very low rate of real return for bonds. That has some implications for retirement planning and bequest motives, no?
As siamond has well observed, the shorter-term performance of bonds (on the scale of decades) has historically been quite volatile, much more so than their longer-term averages on the scale of centuries. That said, there are several factors today working to depress global real interest rates — including aging populations in most advanced countries, low productivity growth, an excess of savings over investment, etc.

In my personal financial planning, I'm resigned to low real bond yields for the foreseeable future. When investors today are pricing 30-year bonds at yields of 1% real or less around the world, it's hard to be more optimistic. So I've adjusted my bond return expectations accordingly — and if actual returns turn out to be higher, so much the better.
Cordially, Todd

GAAP
Posts: 588
Joined: Fri Apr 08, 2016 12:41 pm

Re: The Deep History of Stock & Bond Returns

Post by GAAP » Thu Jun 14, 2018 11:26 am

Is there sufficient data to indicate if the correlation between equity and bonds changed over time?

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: The Deep History of Stock & Bond Returns

Post by SimpleGift » Thu Jun 14, 2018 12:04 pm

GAAP wrote:
Thu Jun 14, 2018 11:26 am
Is there sufficient data to indicate if the correlation between equity and bonds changed over time?
On the scale of centuries, none that I'm aware of. For some reason, those who conduct research into the deep history of asset returns seem to be divided into "bond folks" and "equity folks." Perhaps comparative stock-bond studies exist, but I haven't run across any such research extending back to the 1600s.

On the scale of decades, though, we have good stock-bond correlations going back into the 1800s (at least for the U.S. and U.K. markets). The chart below shows the correlations between U.S. 10-year Treasuries and S&P Composite stocks from 1871.
From this 140-year period, about all we can say is the stock-bond correlation is highly variable over time, changing from positive to negative and then back again. Whether this same correlation pattern might apply in earlier centuries is hard to say.
Last edited by SimpleGift on Thu Jun 14, 2018 12:09 pm, edited 1 time in total.
Cordially, Todd

dbr
Posts: 27207
Joined: Sun Mar 04, 2007 9:50 am

Re: The Deep History of Stock & Bond Returns

Post by dbr » Thu Jun 14, 2018 12:07 pm

I am not sure how to even think about a "correlation" that doesn't correlate even if technically you can compute numbers.

User avatar
siamond
Posts: 4185
Joined: Mon May 28, 2012 5:50 am

Re: The Deep History of Stock & Bond Returns

Post by siamond » Thu Jun 14, 2018 12:17 pm

AtlasShrugged? wrote:
Thu Jun 14, 2018 6:27 am
If I am reading the graph right, it looks like we are about to level off for a long time at a very low rate of real return for bonds. That has some implications for retirement planning and bequest motives, no?
SimpleGift wrote:
Thu Jun 14, 2018 10:19 am
In my personal financial planning, I'm resigned to low real bond yields for the foreseeable future. When investors today are pricing 30-year bonds at yields of 1% real or less around the world, it's hard to be more optimistic. So I've adjusted my bond return expectations accordingly — and if actual returns turn out to be higher, so much the better.
As much as bonds seem to have a rather unpredictable life of their own over decades, I don't disagree with you, and I use similar expectations for my financial planning... Future will tell...

GAAP
Posts: 588
Joined: Fri Apr 08, 2016 12:41 pm

Re: The Deep History of Stock & Bond Returns

Post by GAAP » Thu Jun 14, 2018 12:18 pm

dbr wrote:
Thu Jun 14, 2018 12:07 pm
I am not sure how to even think about a "correlation" that doesn't correlate even if technically you can compute numbers.
I was hoping to be able to evaluate how likely global bonds would be to offer diversification benefits in the future. Market-specific data is readily available -- but doesn't really answer the question for globally diversified portfolio.

GAAP
Posts: 588
Joined: Fri Apr 08, 2016 12:41 pm

Re: The Deep History of Stock & Bond Returns

Post by GAAP » Thu Jun 14, 2018 12:42 pm

siamond wrote:
Thu Jun 14, 2018 12:17 pm
AtlasShrugged? wrote:
Thu Jun 14, 2018 6:27 am
If I am reading the graph right, it looks like we are about to level off for a long time at a very low rate of real return for bonds. That has some implications for retirement planning and bequest motives, no?
SimpleGift wrote:
Thu Jun 14, 2018 10:19 am
In my personal financial planning, I'm resigned to low real bond yields for the foreseeable future. When investors today are pricing 30-year bonds at yields of 1% real or less around the world, it's hard to be more optimistic. So I've adjusted my bond return expectations accordingly — and if actual returns turn out to be higher, so much the better.
As much as bonds seem to have a rather unpredictable life of their own over decades, I don't disagree with you, and I use similar expectations for my financial planning... Future will tell...
I try to estimate a trend using a weighted average of the last 50 years and the post-2000 results. This year, that yields 4.55% for equity, 4.53% for IT bonds, and 0.38% for ST bonds.

The hope is to get the overall trend -- but inform it with more recent results so that my forward look is closer to what ends up happening. I won't know until after the fact, of course...

User avatar
siamond
Posts: 4185
Joined: Mon May 28, 2012 5:50 am

Re: The Deep History of Stock & Bond Returns

Post by siamond » Thu Jun 14, 2018 12:53 pm

GAAP wrote:
Thu Jun 14, 2018 12:42 pm
I try to estimate a trend using a weighted average of the last 50 years and the post-2000 results. This year, that yields 4.55% for equity, 4.53% for IT bonds, and 0.38% for ST bonds.

The hope is to get the overall trend -- but inform it with more recent results so that my forward look is closer to what ends up happening. I won't know until after the fact, of course...
I understand your reasoning, but again, given the erratic history of bonds real returns, I don't think such approach would yield any valuable results (you're making an implicit 'return-to-the-mean' assumption, which was clearly disproved by history).

I agree with SimpleGift here, if I were to make assumptions, I would be VERY conservative for future bonds (real) returns...

GAAP
Posts: 588
Joined: Fri Apr 08, 2016 12:41 pm

Re: The Deep History of Stock & Bond Returns

Post by GAAP » Thu Jun 14, 2018 1:08 pm

siamond wrote:
Thu Jun 14, 2018 12:53 pm
GAAP wrote:
Thu Jun 14, 2018 12:42 pm
I try to estimate a trend using a weighted average of the last 50 years and the post-2000 results. This year, that yields 4.55% for equity, 4.53% for IT bonds, and 0.38% for ST bonds.

The hope is to get the overall trend -- but inform it with more recent results so that my forward look is closer to what ends up happening. I won't know until after the fact, of course...
I understand your reasoning, but again, given the erratic history of bonds real returns, I don't think such approach would yield any valuable results (you're making an implicit 'return-to-the-mean' assumption, which was clearly disproved by history).

I agree with SimpleGift here, if I were to make assumptions, I would be VERY conservative for future bonds (real) returns...
I'm using this trend for a PMT-based withdrawal in a few years. In the meantime, I have to use something to guestimate how RMDs, Roth Conversions, Defined Benefit claiming, etc. will work out. I believe this is similar to longinvest's description of the internal trend for VPW -- I just use a different set of assumptions to SWAG it.

golfCaddy
Posts: 704
Joined: Wed Jan 10, 2018 10:02 pm

Re: The Deep History of Stock & Bond Returns

Post by golfCaddy » Thu Jun 14, 2018 3:38 pm

rudeboy wrote:
Tue Jun 12, 2018 5:25 pm
As you said above, this is the financial equivalent of archeology, and archeologists are able to deduce quite a lot with quite a little. I'm not sure how you couldn't take note of the returns of nine ancient securities that almost perfectly align with the global average returns of the past 100 years.
It gets better. All the data from the 1600s is effectively dominated by the performance of a single company.
Before 1692, our dataset consists of the VOC alone, which was by far the largest company in the Netherlands and the U.K. at that time.
You might think the 1700s was more diversified with data from nine companies, but for most of that time, the index consisted of just five stocks.
The second group of stocks includes the London Assurance Company (LA), the Million Bank (MB), the Royal African Company (RAC), and the Royal Exchange Assurance Company (REA). ...
Data coverage is more limited for the second group of companies. In general, prices go back to the initial issuance of each security, but coverage ends in 1734 when the Course of Exchange stops reporting their prices (Neal 1990).

GAAP
Posts: 588
Joined: Fri Apr 08, 2016 12:41 pm

Re: The Deep History of Stock & Bond Returns

Post by GAAP » Fri Jun 15, 2018 11:14 am

This is what stands out to me so far:
  1. Bonds show a longterm downward trend with no mean reversion. Now at 2% Real, with nothing to indicate they won't go lower.
  2. Stocks show a longterm mean reversion trend around 5-6% Real, with nothing to indicate that will change.
  3. In any individual country, stock growth is tied to GDP growth.
  4. Stocks/Bonds correlation is highly variable, surprisingly positive at times -- and more so on shorter terms (1 year vs. 5 years)
It's hard for me to believe that Bond returns will go much lower. I wonder how much the degree of debt-financed business growth is directly tied to this long term downward trend -- and how much the trend is tied to the growth in debt financing. Clearly, there has to be a bottom somewhere, but where?

I'm having much more difficulty with the concept of bonds as "safe". If they have a strong positive correlation with stocks during crashes, they provide no real safety -- but they can certainly act as portfolio drag. I see the most potential value for those who use a V-shaped AA during some period around a retirement date.

If stock growth is tied to GDP growth, then maybe we should be forecasting which economies will grow the most. Since that is difficult and unreliable at best, I find global equity indexing to be the best alternative. Essentially, it provides "economy diversification", not just equity diversification.

Post Reply