T'was the Worst of Times — Case Study of 1973-1985

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
SimpleGift
Posts: 4165
Joined: Tue Feb 08, 2011 3:45 pm

T'was the Worst of Times — Case Study of 1973-1985

Post by SimpleGift »

This post looks at the behavior of investment portfolios with various stock-bond mixes during the double-digit inflation episodes of the 1970s, starting at the stock market peak in January 1973. Let's try to focus any discussion on the investing aspects of this period, and avoid the economic and monetary policy aspects — since experts are still arguing over this history even today!

Inflation had been low for most of the 1960s, and was even below 3% in 1972. However, the OPEC oil embargo of 1973-74 set off an inflationary spiral that soon reached double-digits. Initially, stock-heavy portfolios were hit the hardest, declining up to 50% real within 18 months, to the end of 1974 (chart below). Bonds losses were 16% real during this period. Note that portfolio performance is real (inflation-adjusted), total returns.

In 1979, the Iranian Revolution delivered another oil supply shock, leading to a second round of double-digit inflation. As inflation accelerated, bond-heavy portfolios suffered the most, declining up to 40% real by 1981. Once inflation subsided, all of the various stock/bond portfolios regained their previous real values more or less together in 1985 — after 12 years of zero real portfolio growth.
  • Image
    NOTE: Stocks are S&P 500, bonds are 10-year Treasuries.
    Sources: Monthly S&P stock returns from Shiller; monthly 10-year Treasury returns from Medium
None of this is to suggest that the current market and economic climate is like the 1970s, or that we are likely to experience a repeat of those times. However, it can be instructive to observe the behavior of various stock-bond portfolios under rapid inflation — as much as the example of the 1970s can offer to us.

Thoughts?
Last edited by SimpleGift on Tue May 10, 2022 9:14 pm, edited 3 times in total.
Marseille07
Posts: 9241
Joined: Fri Nov 06, 2020 1:41 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by Marseille07 »

Excellent post.

Personally I'm surprised by the poor performance by 100/0 here. People often say equities beat inflation but that doesn't appear to hold water here.
85% US + FM + RSU | 15% Cash
User avatar
Topic Author
SimpleGift
Posts: 4165
Joined: Tue Feb 08, 2011 3:45 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by SimpleGift »

Marseille07 wrote: Tue May 10, 2022 5:16 pm People often say equities beat inflation but that doesn't appear to hold water here.
From what I've read, the stock market was more heavily weighted toward industrial companies in the 1970s, and had largely unionized labor forces. As inflation spiraled higher, unions were able to build in generous cost-of-living adjustments into their contracts. Companies gradually saw their profit margins shrink, as unit labor costs increased much faster than they could be passed on to customers.

All this is a far cry from the tech-dominated stock market of today, with weak or non-existent labor organization. Whether this makes today's companies less susceptible to inflation pressures is hard to say.
Last edited by SimpleGift on Tue May 10, 2022 5:53 pm, edited 1 time in total.
Marseille07
Posts: 9241
Joined: Fri Nov 06, 2020 1:41 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by Marseille07 »

SimpleGift wrote: Tue May 10, 2022 5:49 pm
Marseille07 wrote: Tue May 10, 2022 5:16 pm People often say equities beat inflation but that doesn't appear to hold water here.
From what I've read, the stock market was more heavily weighted toward industrial companies in the 1970s, and had largely unionized labor forces. As inflation spiraled higher, unions were able to build in generous cost-of-living adjustments into their contracts. Companies gradually saw their profit margins shrink, as unit labor costs increased much faster than they could be passed on to customers.

All this is a far cry from the tech-dominated stock market of today, with weak or non-existent labor organization.
Thank you. I don't have first-hand experience but what you're describing sounds accurate to me :beer
85% US + FM + RSU | 15% Cash
frugalecon
Posts: 549
Joined: Fri Dec 05, 2014 12:39 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by frugalecon »

Thanks for posting. While this suggests that it was a very challenging scenario for someone who retired at the beginning of the period, for those who were buying more shares along the way during the 1970s it was a setup for very nice returns down the pike. I have been a bit disappointed that I am not earlier in my accumulation phase during the recent market downturn. (Though I had turned down my stock purchases in my retirement plan over the last year, and increased them recently. Yes, I have engaged in mild market timing.)
Tinyz
Posts: 102
Joined: Fri Nov 19, 2021 9:59 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by Tinyz »

If history is ever a guide, it takes 2 years to recover from oil shock. As I continuously look at the active rig for the past few months, there seems to be a weird trend going on there. I don't know why but I don't feel good. :(
User avatar
arcticpineapplecorp.
Posts: 10244
Joined: Tue Mar 06, 2012 9:22 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by arcticpineapplecorp. »

Marseille07 wrote: Tue May 10, 2022 5:16 pm Excellent post.

Personally I'm surprised by the poor performance by 100/0 here. People often say equities beat inflation but that doesn't appear to hold water here.
Nothing, not any of the allocations performed well against inflation. But it looks to me that starting in 1980 the 100/0 portfolio started outperforming the less risky portfolios. So about 7 years to start out performing. Thats no different than other bear markets i've seen.
It's hard to accept the truth when the lies were exactly what you wanted to hear. Investing is simple, but not easy. Buy, hold & rebalance low cost index funds & manage taxable events. Asking Portfolio Questions | Wiki
Hola
Posts: 15
Joined: Thu Nov 04, 2021 2:40 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by Hola »

Tinyz wrote: Tue May 10, 2022 6:28 pm If history is ever a guide, it takes 2 years to recover from oil shock. As I continuously look at the active rig for the past few months, there seems to be a weird trend going on there. I don't know why but I don't feel good. :(
Do you want to share more about the trend you’re seeing? Thx.
000
Posts: 7631
Joined: Thu Jul 23, 2020 12:04 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by 000 »

TIPS: The Investment Poorly Substituted
User avatar
9-5 Suited
Posts: 938
Joined: Thu Jun 23, 2016 12:14 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by 9-5 Suited »

frugalecon wrote: Tue May 10, 2022 6:12 pm Thanks for posting. While this suggests that it was a very challenging scenario for someone who retired at the beginning of the period, for those who were buying more shares along the way during the 1970s it was a setup for very nice returns down the pike.
Yeah if you were earning a nice salary in the 70s and investing it wisely, you are what’s known as “very rich”. Or you heirs are, whatever the case may be. Always two sides to every coin, since most of these discussions focus on the impact to retirees.
User avatar
Topic Author
SimpleGift
Posts: 4165
Joined: Tue Feb 08, 2011 3:45 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by SimpleGift »

Looking at real earnings growth during the 1973-1985 period, it's not a surprise that there was no real stock market appreciation during the period. Real earnings growth was essentially zero over 15 years (chart below).
  • Image
    NOTE: Inflation numbers are average over the sub-period. Data source: S&P 500 earnings from Shiller.
It will be interesting to see how S&P 500 earnings hold up in the quarters ahead, for the rest of 2022, especially if inflation numbers remain elevated.
Last edited by SimpleGift on Tue May 10, 2022 7:40 pm, edited 1 time in total.
MarkRoulo
Posts: 700
Joined: Mon Jun 22, 2015 10:25 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by MarkRoulo »

Marseille07 wrote: Tue May 10, 2022 5:16 pm Excellent post.

Personally I'm surprised by the poor performance by 100/0 here. People often say equities beat inflation but that doesn't appear to hold water here.
Equities perform worse than typical if one starts at a market peak. The OP started with a market peak for this illustration.
longinvest
Posts: 4995
Joined: Sat Aug 11, 2012 8:44 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by longinvest »

SimpleGift, something seems off in your chart. Are-you doubly-counting inflation?

Here's a copy of an earlier post I made:
longinvest wrote: Sun Mar 27, 2022 5:44 pm Inflation jumped to 8.7% in 1973 and went further up to 12.3% in 1974. In 1979, inflation went as high as 13.3%. It didn't get below 4% until 1982. Here's a growth chart of stocks, bonds, and inflation from 1973 to 1985:

Image

During the entire period, intermediate bonds had positive total returns every year. In contrast, the S&P 500 (with dividends reinvested) fluctuated. From 1973 to 1975 it lost approximatively -30% (cumulative) while inflation was going up by more than 20% (cumulative).

Note that inflation in the chart represents the growth of the Consumer Price Index for All Urban Consumers (CPI-U).
All-in-one global balanced index ETF | Variable Percentage Withdrawal (VPW) https://www.bogleheads.org/wiki/Variable_percentage_withdrawal#VPW_Accumulation_And_Retirement_Worksheet
User avatar
Topic Author
SimpleGift
Posts: 4165
Joined: Tue Feb 08, 2011 3:45 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by SimpleGift »

longinvest wrote: Tue May 10, 2022 7:42 pm SimpleGift, something seems off in your chart.
The portfolio performances in the OP chart are real (inflation-adjusted), total returns. The chart you posted appears to be the nominal growth of stocks and bonds during the period, if I am not mistaken.
longinvest
Posts: 4995
Joined: Sat Aug 11, 2012 8:44 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by longinvest »

SimpleGift wrote: Tue May 10, 2022 7:57 pm The portfolio performances in the OP chart are real (inflation-adjusted), total returns.
SimpleGift, this confirms that your chart doubly counts inflation.
All-in-one global balanced index ETF | Variable Percentage Withdrawal (VPW) https://www.bogleheads.org/wiki/Variable_percentage_withdrawal#VPW_Accumulation_And_Retirement_Worksheet
User avatar
Topic Author
SimpleGift
Posts: 4165
Joined: Tue Feb 08, 2011 3:45 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by SimpleGift »

longinvest wrote: Tue May 10, 2022 8:00 pm SimpleGift, this confirms that your chart doubly counts inflation.
Not sure I understand. The dashed yellow inflation line in your chart is identical to the black inflation line in the OP. The monthly inflation data, plus the monthly stock return data in the OP, are directly from Shiller's database.
User avatar
nisiprius
Advisory Board
Posts: 46543
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by nisiprius »

Marseille07 wrote: Tue May 10, 2022 5:16 pm Excellent post.

Personally I'm surprised by the poor performance by 100/0 here. People often say equities beat inflation but that doesn't appear to hold water here.
We should assemble a list of things that aren't so, but people often say them. This is a good case in point. In 1973--1973!--Benjamin Graham (Warren Buffett's guru) wrote
Benjamin Graham, on p. 20 of 'The Intelligent Investor,' 1973 edition wrote:On this point we can be categorical. There is no close time connection between inflationary (or deflationary) conditions and the movement of common-stock earnings and prices. The obvious example is the recent period 1966-1970. The rise in the cost of living was 22%... but both stock earnings and stock prices have declined since 1965. There are similar contradictions in both directions in the record of previous five-year periods.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
nisiprius
Advisory Board
Posts: 46543
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by nisiprius »

SimpleGift wrote: Tue May 10, 2022 8:07 pm
longinvest wrote: Tue May 10, 2022 8:00 pm SimpleGift, this confirms that your chart doubly counts inflation.
Not sure I understand. The dashed yellow inflation line in your chart is identical to the black inflation line in the OP. The monthly inflation data, plus the monthly stock return data in the OP, are directly from Shiller's database.
If everything is corrected for inflation, then the curve for "inflation" is the level horizontal dotted line at 0%. The portfolios are falling short of inflation when they are below the level horizontal dotted line. When they have a positive slope they are winning against inflation regardless of what the "inflation" line is doing.

The portfolio curves "look right" because I know from memory that stock market real returns from 1966-1982 inclusive were almost zero, very slightly negative.

The curve labeled "inflation" is not, itself, inflation-corrected, so it is not being plotted consistently on the same scale as the portfolio curves. It's also confusing because the curve seems to be plotting the CPI itself, the cumulative effect, while the word "inflation" often refers to the annual change in CPI. I would suggest that the curve be labeled "CPI" rather than inflation, that it not be adjusted to 1.0, and that it be plotted on a second vertical axis.

The visual impression is very confusing if we gauge the portfolio curves according to their relationship to the curve labeled "inflation."
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
HootingSloth
Posts: 796
Joined: Mon Jan 28, 2019 3:38 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by HootingSloth »

nisiprius wrote: Tue May 10, 2022 8:17 pm
SimpleGift wrote: Tue May 10, 2022 8:07 pm
longinvest wrote: Tue May 10, 2022 8:00 pm SimpleGift, this confirms that your chart doubly counts inflation.
Not sure I understand. The dashed yellow inflation line in your chart is identical to the black inflation line in the OP. The monthly inflation data, plus the monthly stock return data in the OP, are directly from Shiller's database.
If everything is corrected for inflation, then the curve for "inflation" is the level horizontal dotted line at 0%. The portfolios are falling short of inflation when they are below the level horizontal dotted line. When they have a positive slope they are winning against inflation regardless of what the "inflation" line is doing.

The portfolio curves "look right" because I know from memory that stock market real returns from 1966-1982 inclusive were almost zero, very slightly negative.

The curve labeled "inflation" is not, itself, inflation-corrected, so it is not being plotted consistently on the same scale as the portfolio curves. It's also confusing because the curve seems to be plotting the CPI itself, the cumulative effect, while the word "inflation" often refers to the annual change in CPI. I would suggest that the curve be labeled "CPI" rather than inflation, that it not be adjusted to 1.0, and that it be plotted on a second vertical axis.

The visual impression is very confusing if we gauge the portfolio curves according to their relationship to the curve labeled "inflation."
The intent of the original graph seemed clear enough to me. The line labeled inflation is charted against a different vertical axis than the portfolio lines. It is not uncommon to show the relationship between two trends by putting them on the same graph with two different vertical axes. If the inflation line were charted on the same vertical axis, it would just be a horizontal line, as you say. This would make it impossible to see the relationship (which periods have higher inflation and which lower) that SimpleGift was trying to illustrate. Perhaps clearer labeling would help in avoiding the understandable confusion though.
Global Market Portfolio + modest tilt towards volatility (80/20->60/40 as approach FI) + modest tilt away from exchange rate risk (80% global+20% U.S. stocks; currency-hedge bonds) + tax optimization
000
Posts: 7631
Joined: Thu Jul 23, 2020 12:04 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by 000 »

It would be clearer to remove the black line and just annotate the real return chart with the CPI percentages at various points.
User avatar
Topic Author
SimpleGift
Posts: 4165
Joined: Tue Feb 08, 2011 3:45 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by SimpleGift »

nisiprius wrote: Tue May 10, 2022 8:17 pm The curve labeled "inflation" is not, itself, inflation-corrected, so it is not being plotted consistently on the same scale as the portfolio curves.
Ah, yes, I see the point of confusion. I've re-posted the chart in the OP without the inflation curve, to focus on the portfolio performances.

Thank you for the clarification.
User avatar
nisiprius
Advisory Board
Posts: 46543
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by nisiprius »

Using only some spreadsheet tools I already have at hand, I don't have an easy way to plot stock+bond portfolios, nor an easy way to plot more than three curves, but here are the inflation-adjusted growth charts, starting at 1966, for the Ibbotson Associates SBBI series for large-company stocks (grey), intermediate-term government bonds (yellow), and long-term government bonds (blue).

It's just a confirmation of the general accuracy of your data.

Image

And here I just used intermediate bonds and adjusted the colors and axis ranges a bit to make it look similar to yours...

Image
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
000
Posts: 7631
Joined: Thu Jul 23, 2020 12:04 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by 000 »

According to PV, the ten year had a correlation with the stock market of:
+0.30 from 1973 - 1985
-0.11 from 1986 - 2020
+0.36 from 2021 - April 2022

Conclusion: inflation matters, though rebalancing stocks and bonds may still have helped somewhat.
Last edited by 000 on Tue May 10, 2022 9:12 pm, edited 1 time in total.
User avatar
Topic Author
SimpleGift
Posts: 4165
Joined: Tue Feb 08, 2011 3:45 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by SimpleGift »

000 wrote: Tue May 10, 2022 8:29 pm It would be clearer to remove the black line and just annotate the real return chart with the CPI percentages at various points.
Good suggestion. The OP chart now has annotations showing the annualized inflation rates for various sub-periods.
000
Posts: 7631
Joined: Thu Jul 23, 2020 12:04 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by 000 »

SimpleGift wrote: Tue May 10, 2022 9:09 pm
000 wrote: Tue May 10, 2022 8:29 pm It would be clearer to remove the black line and just annotate the real return chart with the CPI percentages at various points.
Good suggestion. The OP chart now has annotations showing the annualized inflation rates for various sub-periods.
Nice! :sharebeer
JBTX
Posts: 9281
Joined: Wed Jul 26, 2017 12:46 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by JBTX »

Marseille07 wrote: Tue May 10, 2022 5:16 pm Excellent post.

Personally I'm surprised by the poor performance by 100/0 here. People often say equities beat inflation but that doesn't appear to hold water here.
It is interesting how people are finally taking note of inflation. There has been a long running narrative of stocks holding up to inflation. Perhaps modest inflation, and if you wait long enough, yes stocks will eventually catch up, but a significantly higher than average expected run of inflation hurts both stocks and bonds.
User avatar
Topic Author
SimpleGift
Posts: 4165
Joined: Tue Feb 08, 2011 3:45 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by SimpleGift »

nisiprius wrote: Tue May 10, 2022 8:48 pm It's just a confirmation of the general accuracy of your data.
Thank you, nisi. Appreciate your taking the time to check the real stock and bond returns for the period.
longinvest
Posts: 4995
Joined: Sat Aug 11, 2012 8:44 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by longinvest »

SimpleGift wrote: Tue May 10, 2022 8:07 pm
longinvest wrote: Tue May 10, 2022 8:00 pm SimpleGift, this confirms that your chart doubly counts inflation.
Not sure I understand. The dashed yellow inflation line in your chart is identical to the black inflation line in the OP. The monthly inflation data, plus the monthly stock return data in the OP, are directly from Shiller's database.
nisiprius wrote: Tue May 10, 2022 8:17 pm If everything is corrected for inflation, then the curve for "inflation" is the level horizontal dotted line at 0%. The portfolios are falling short of inflation when they are below the level horizontal dotted line. When they have a positive slope they are winning against inflation regardless of what the "inflation" line is doing.

The portfolio curves "look right" because I know from memory that stock market real returns from 1966-1982 inclusive were almost zero, very slightly negative.

The curve labeled "inflation" is not, itself, inflation-corrected, so it is not being plotted consistently on the same scale as the portfolio curves. It's also confusing because the curve seems to be plotting the CPI itself, the cumulative effect, while the word "inflation" often refers to the annual change in CPI. I would suggest that the curve be labeled "CPI" rather than inflation, that it not be adjusted to 1.0, and that it be plotted on a second vertical axis.

The visual impression is very confusing if we gauge the portfolio curves according to their relationship to the curve labeled "inflation."
SimpleGift, as explained by Nisiprius, the original chart of the OP, before modification, was confusing. Its increasing inflation line doubly counted inflation relative to the inflation-adjusted growth lines of stocks and bonds. The modified chart in the OP is clearer.
All-in-one global balanced index ETF | Variable Percentage Withdrawal (VPW) https://www.bogleheads.org/wiki/Variable_percentage_withdrawal#VPW_Accumulation_And_Retirement_Worksheet
Nathan Drake
Posts: 3323
Joined: Mon Apr 11, 2011 12:28 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by Nathan Drake »

Now plot SCV and exUS against this period
20% VOO | 30% VXUS | 20% AVUV | 15% AVDV | 15% AVES
Booglie
Posts: 133
Joined: Sun Aug 15, 2021 2:17 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by Booglie »

If anything, today has the potential to be worse due to an aging population and world economy deceleration.

Also, weak unionizing works both ways. Because workers are in a scenario where their jobs are being automated, this pressures against wage adjustments. If wages don't rise but inflation does, it means that people's purchasing power will diminish, and consumption will decelerate further.

In other words, if no one can afford anything because everyone's unemployed, who will support the economy?

On the top of THAT, we have COVID (which has not totally gone away) and war.
User avatar
Topic Author
SimpleGift
Posts: 4165
Joined: Tue Feb 08, 2011 3:45 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by SimpleGift »

Nathan Drake wrote: Tue May 10, 2022 11:27 pm Now plot SCV and exUS against this period.
Was able to find returns for some common asset classes during the 1972-1981 decade (table below). These come from Portfolio Visualizer data, Simba spreadsheet data, with the rental housing returns from the JST database. All returns are average annual real (inflation-adjusted), total returns for the decade:
  • Image
The only returns I would flag for comment are the gold returns. The 1972-1981 decade saw a huge run-up in real gold prices as gold transitioned from an artificially-low pegged USD price under Bretton Woods to a free-floating price. That's not going to happen again, so real gold returns in this period are abnormal and hardly representative.

Clearly, S&P 500 stocks and nominal bonds were not a winning combination during this decade of 9% average annual inflation.
Last edited by SimpleGift on Wed May 11, 2022 7:53 am, edited 2 times in total.
User avatar
vineviz
Posts: 11576
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by vineviz »

SimpleGift wrote: Wed May 11, 2022 6:47 am Clearly, S&P 500 stocks and nominal bonds were not a winning combination during this decade of 9% average annual inflation.
Just for reference, here are plots of the inflation adjusted growth of international stocks and small cap value US stocks compared to large cap US stocks for the 1973-1989 period.

Image

For the sub-period of 1973 to 1979, an investment in large cap US stocks lost 31% of its purchasing power. International stocks lost 4% and small cap value US stocks lost 2%.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
halfnine
Posts: 1869
Joined: Tue Dec 21, 2010 1:48 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by halfnine »

SimpleGift wrote: Wed May 11, 2022 6:47 am
Nathan Drake wrote: Tue May 10, 2022 11:27 pm Now plot SCV and exUS against this period.
Was able to find returns for some common asset classes during the 1972-1981 decade (table below). These come from Portfolio Visualizer data, Simba spreadsheet data, with the rental housing returns from the JST database. All returns are average annual real (inflation-adjusted), total returns for the decade:
  • Image
The only returns I would flag for comment are the gold returns. The 1972-1981 decade saw a huge run-up in real gold prices as gold transitioned from an artificially-low pegged USD price under Bretton Woods to a free-floating price. That's not going to happen again, so real gold returns in this period are abnormal and hardly representative.

Clearly, S&P 500 stocks and nominal bonds were not a winning combination during this decade of 9% average annual inflation.
From 1968 to 1973 the price of gold (black market rates) likely accurately reflected the collapse of Bretton Woods. From 1973 to 1980 rate of change in gold prices are fairly well correlated with the rate of change in CPI. Based on any data I have seen there is nothing to indicate that the unwinding of Bretton Woods persisted beyond 1973 as a major driving force for the pricing of gold.
martincmartin
Posts: 254
Joined: Wed Jul 02, 2014 3:04 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by martincmartin »

It turns out, the worst starting year for a 30 year retirement isn't 1973, but 1966. 1968 is second worst. So it might be interesting to extend your charts back to then.
Valuethinker
Posts: 44460
Joined: Fri May 11, 2007 11:07 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by Valuethinker »

Booglie wrote: Wed May 11, 2022 5:03 am If anything, today has the potential to be worse due to an aging population and world economy deceleration.

Also, weak unionizing works both ways. Because workers are in a scenario where their jobs are being automated, this pressures against wage adjustments. If wages don't rise but inflation does, it means that people's purchasing power will diminish, and consumption will decelerate further.

In other words, if no one can afford anything because everyone's unemployed, who will support the economy?

On the top of THAT, we have COVID (which has not totally gone away) and war.
The hypothesis of high unemployment does not seem to be validated, at least at the moment?

Rather, the developed countries have had low birth rates for decades. The US was substantially ahead but I believe it has faded back in recent years.

Thus, age & illness (exacerbated by Covid) is creating exit from the labour force. The workforces of developed countries are not growing. The US Labour Market, for example, is as close to full employment as it has been in 30+ years?

At the same time, most developed countries have placed greater restrictions in immigration. How well these will hold in the long term is a moot point, but it was certainly, for example, a big part of the Brexit vote - Britain had experienced a wave of immigration from the accession states (in particular Poland and Lithuania) in the 10 years or so before the 2016 vote.

As the average age of the world increases, and it is going to do so quite rapidly, that speaks to a labour force shortage.

(The only countries where birth rates have remained high are parts of the Middle East, and sub Saharan Africa. It's hard to see a world where Europe, say, accepts many millions of Africans -- but of course we shall see).

Yes there will be more "AI" and robotics: the Japanese are really pushing the frontier on this, because they are confronting a declining population.

But at least for the next few years, it seems likely we will have too few workers rather than too many?
Valuethinker
Posts: 44460
Joined: Fri May 11, 2007 11:07 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by Valuethinker »

Tinyz wrote: Tue May 10, 2022 6:28 pm If history is ever a guide, it takes 2 years to recover from oil shock. As I continuously look at the active rig for the past few months, there seems to be a weird trend going on there. I don't know why but I don't feel good. :(
If you look to the mid 1970s or 1979-80 I believe recovery took a lot longer? Depending on how you define "recovery"?

US oil consumption did not cross its pre 1979 level again until the mid-late 80s? There was a big shift away from oil in electric power generation, and towards more fuel efficient vehicles etc. Energy conservation became a serious objective for the first time in American history.

What is true is that oil from fracking is much more responsive to the oil price in the short term than it was. Previously new resources were more like Brazilian offshore - many years to find, many billions of dollars to develop to exploitation. At least that has been the pattern since about 2010?

So on the sluggishness of drilling activity:

- the fracking industry consumed over $1 trillion of debt + equity (I believe is the number) in the previous boom. Producing mediocre returns for investors. There is much greater evidence of capital discipline on the part of the companies this time. Reflecting in part perhaps the entry of the oil majors into the industry - who have a record of discipline in deploying capital over decades (having got it badly wrong in the 1970s). Exxon-Mobil in particular (but not only).

Wall Street is just not underwriting every fracking play going. Particularly not with rising interest rates generally & a lot of uncertainty. If the war with Russia-Ukraine ends, or OPEC decides to open up the taps, the price could go right back down.

- as with so many other industries, there are supply issues. Workers have left the industry (stereotypically, to go work at Amazon Distribution Centers). Equipment is not available. Everybody is really scrambling, and costs are going up, which is making everyone cautious. But it's also just not possible to find the people and equipment to drill.

- there's an open question whether the fracking boom was largely the product of 1-2 major shale formations and that is not repeated geologically elsewhere. Production from newly fracked wells falls relatively quickly, so holding or increasing output is dependent upon a *lot* of new drilling activity.

I should stress that I am no expert, but from what I read.
make_a_better_world
Posts: 287
Joined: Mon Mar 18, 2013 11:55 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by make_a_better_world »

This is what happened to real estate over the same period. Well at least home prices:

Image

https://archive.curbed.com/2018/4/10/17 ... nt-gi-bill

Image

https://malibutimes.com/article_d8fd28f ... 581abe6e4d
alluringreality
Posts: 693
Joined: Tue Nov 12, 2019 10:59 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by alluringreality »

SimpleGift wrote: Wed May 11, 2022 6:47 am
Nathan Drake wrote: Tue May 10, 2022 11:27 pm Now plot SCV and exUS against this period.
Was able to find returns for some common asset classes during the 1972-1981 decade (table below). These come from Portfolio Visualizer data, Simba spreadsheet data, with the rental housing returns from the JST database. All returns are average annual real (inflation-adjusted), total returns for the decade:
  • Image
The only returns I would flag for comment are the gold returns. The 1972-1981 decade saw a huge run-up in real gold prices as gold transitioned from an artificially-low pegged USD price under Bretton Woods to a free-floating price. That's not going to happen again, so real gold returns in this period are abnormal and hardly representative.

Clearly, S&P 500 stocks and nominal bonds were not a winning combination during this decade of 9% average annual inflation.
For seeming to only have a difference of maybe one year, I'm a bit surprised by some of the differences compared to the Global Financial Data chart from Global Asset Allocation by Mebane Faber. It looks like 1972 was a good year for lots of categories, but I'll presume some of the differences may be due to data sources.

Real Returns 1973-1981:
Gold 13.02%
Foreign EEM Stock 4.29%
GSCI 3.27%
US Small Stock 1.43%
TIPS synthetic series 0.49%
T-Bills -0.71%
Foreign EAFE Stock -1.22%
REITs -2.28%
Foreign 10 Year Bonds -2.66%
Corporate Bonds -3.79%
US Large Stock -3.92%
10 Year US Bonds -5.08%
30 Year US Bonds -8.65%
30% Savings Bonds, 45% US Indexes, 25% Ex-US Indexes
Valuethinker
Posts: 44460
Joined: Fri May 11, 2007 11:07 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by Valuethinker »

make_a_better_world wrote: Thu May 12, 2022 9:40 am This is what happened to real estate over the same period. Well at least home prices:

Image

https://archive.curbed.com/2018/4/10/17 ... nt-gi-bill

Image

https://malibutimes.com/article_d8fd28f ... 581abe6e4d
Real estate is a considerable puzzle.

There is such wide dispersion between US cities in the last 50 years that it's dangerous to generalize.

However broadly it appears there was a structural change in about the early 1970s:

- housing in downtown/ midtown areas started to become more valuable (having fallen relative to the new suburbs) due to changing demographics, distaste for commuting, increasing traffic volume (shorter time to work), changing preferences - I recognise this wasn't really visible until the 1980s in many cities, and not until into the 2000s generally (if at all - many cities like Detroit it just didn't happen)

If you think what Brooklyn has done, say, since 1970, it's pretty incredible, at least in some parts (if not all). From declining working class city to heart of NY gentrification.

- local opposition to more intense development aka "NIMBY" restrictions became stronger and more politically influential. Thus many of the high demand areas of the US (coastal California, Portland & Seattle, Boston-Washington) became much more expensive in real terms. Contrast Chicago housing prices to New York ones, for example. Atlanta to NE cities. Las Vegas to SoCal. etc.


Added to that you had plunging real interest rates, which generally is a good thing for real estate prices.

The 2001-06ish housing bubble and subsequent collapse obscured some of these trends, but didn't entirely eliminate them.
User avatar
Topic Author
SimpleGift
Posts: 4165
Joined: Tue Feb 08, 2011 3:45 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by SimpleGift »

halfnine wrote: Wed May 11, 2022 1:28 pm From 1968 to 1973 the price of gold (black market rates) likely accurately reflected the collapse of Bretton Woods. From 1973 to 1980 rate of change in gold prices are fairly well correlated with the rate of change in CPI. Based on any data I have seen there is nothing to indicate that the unwinding of Bretton Woods persisted beyond 1973 as a major driving force for the pricing of gold.
You could be right. Not sure there is any way to disentangle the unwinding of Bretton Woods from the impact of double-digit inflation on the price of gold during the 1970s. My comment comes from looking at the real (inflation-adjusted) price of gold by decade, since 1900:
On the scale of a century, gold has kept up with inflation — but the real return of gold during the decade of the 1970s does stand out as a bit abnormal.
User avatar
Topic Author
SimpleGift
Posts: 4165
Joined: Tue Feb 08, 2011 3:45 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by SimpleGift »

alluringreality wrote: Thu May 12, 2022 9:58 am For seeming to only have a difference of maybe one year, I'm a bit surprised by some of the differences compared to the Global Financial Data chart from Global Asset Allocation by Mebane Faber. It looks like 1972 was a good year for lots of categories, but I'll presume some of the differences may be due to data sources.
Yes, sometime it would be interesting to do a Forum post on the differing historical returns, by asset class, which are derived from various data sources. There can be quite a variance in the historical record.

Suffice to say, the double-digit inflation era of the 1970s was not kind to most financial assets.
Booglie
Posts: 133
Joined: Sun Aug 15, 2021 2:17 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by Booglie »

Valuethinker wrote: Thu May 12, 2022 9:16 am
Booglie wrote: Wed May 11, 2022 5:03 am If anything, today has the potential to be worse due to an aging population and world economy deceleration.

Also, weak unionizing works both ways. Because workers are in a scenario where their jobs are being automated, this pressures against wage adjustments. If wages don't rise but inflation does, it means that people's purchasing power will diminish, and consumption will decelerate further.

In other words, if no one can afford anything because everyone's unemployed, who will support the economy?

On the top of THAT, we have COVID (which has not totally gone away) and war.
The hypothesis of high unemployment does not seem to be validated, at least at the moment?

Rather, the developed countries have had low birth rates for decades. The US was substantially ahead but I believe it has faded back in recent years.
While unemployment rates in the US are at record lows, the rest of the world is not experiencing the same fate. Eventually, if the world as a whole slows down, it's very likely this will pressure the US too.
e
As for automation, it is being seen as a response to an aging population. But the core problem with automation is that its main goal is to do more with less. At first, this only eliminated low-paying jobs; but now, with neural technologies, we have AIs that can write, paint, draw, and even translate audio (Google is releasing such a feature today).

In order for unemployment to rise, machines don't have to do everything better than us. All it takes is they do most jobs more efficiently and cheaply than we do. Our economy was designed by people, for people.

What happens when robots can produce all our wealth for "free", but no one can afford it? This is a question without a good answer. In such a system, our current capitalistic system would have to fall. It wouldn't necessarily give room to socialism, but what we could see is a system that is artificially stimulated.

We saw a taste of that artificially stimulated system, if only for different reasons, when COVID struck. But if eventually there are no jobs, this may be a new normal.

The challenge with the next decades is that this automation won't be gone; it'll be evermore efficient.

On the short and medium term, we are reaping the consequences of cutting costs: because companies are always trying to reduce wages and moving their productions to other countries, not only industrial production is concentrated at a few countries, but this also threatens specialized labor. Why would one invest on specializing if their profession is seen more and more like a commodity? Why study for several years and acquire a huge debt in education if companies are willing to pay less and less for it?

This is one of the reason some fields lack qualified people.
alfaspider
Posts: 4093
Joined: Wed Sep 09, 2015 4:44 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by alfaspider »

Booglie wrote: Thu May 12, 2022 11:43 am
Valuethinker wrote: Thu May 12, 2022 9:16 am
Booglie wrote: Wed May 11, 2022 5:03 am If anything, today has the potential to be worse due to an aging population and world economy deceleration.

Also, weak unionizing works both ways. Because workers are in a scenario where their jobs are being automated, this pressures against wage adjustments. If wages don't rise but inflation does, it means that people's purchasing power will diminish, and consumption will decelerate further.

In other words, if no one can afford anything because everyone's unemployed, who will support the economy?

On the top of THAT, we have COVID (which has not totally gone away) and war.
The hypothesis of high unemployment does not seem to be validated, at least at the moment?

Rather, the developed countries have had low birth rates for decades. The US was substantially ahead but I believe it has faded back in recent years.
While unemployment rates in the US are at record lows, the rest of the world is not experiencing the same fate. Eventually, if the world as a whole slows down, it's very likely this will pressure the US too.
e
As for automation, it is being seen as a response to an aging population. But the core problem with automation is that its main goal is to do more with less. At first, this only eliminated low-paying jobs; but now, with neural technologies, we have AIs that can write, paint, draw, and even translate audio (Google is releasing such a feature today).

In order for unemployment to rise, machines don't have to do everything better than us. All it takes is they do most jobs more efficiently and cheaply than we do. Our economy was designed by people, for people.

What happens when robots can produce all our wealth for "free", but no one can afford it? This is a question without a good answer. In such a system, our current capitalistic system would have to fall. It wouldn't necessarily give room to socialism, but what we could see is a system that is artificially stimulated.

We saw a taste of that artificially stimulated system, if only for different reasons, when COVID struck. But if eventually there are no jobs, this may be a new normal.

The challenge with the next decades is that this automation won't be gone; it'll be evermore efficient.

On the short and medium term, we are reaping the consequences of cutting costs: because companies are always trying to reduce wages and moving their productions to other countries, not only industrial production is concentrated at a few countries, but this also threatens specialized labor. Why would one invest on specializing if their profession is seen more and more like a commodity? Why study for several years and acquire a huge debt in education if companies are willing to pay less and less for it?

This is one of the reason some fields lack qualified people.
I think even perfect automation still creates a huge number of jobs for people looking to improve machines and tell them what to do. It also expands the overall universe of human activities. I'd look at accounting for example of what automation does. Most companies used to have tons of people whose job it was to crunch numbers for the books. Computers have made that obsolete. We even have software that automates a lot of accounting process. But somehow, most companies still have a similar number of accountants relative to other employees. Why? As the basics got automated, the scope of what accountants were able to track (and what they were required to track) grew along with it. I think you will find that happen in a lot of different fields.
Booglie
Posts: 133
Joined: Sun Aug 15, 2021 2:17 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by Booglie »

alfaspider wrote: Thu May 12, 2022 12:19 pm
Booglie wrote: Thu May 12, 2022 11:43 am
Valuethinker wrote: Thu May 12, 2022 9:16 am
Booglie wrote: Wed May 11, 2022 5:03 am If anything, today has the potential to be worse due to an aging population and world economy deceleration.

Also, weak unionizing works both ways. Because workers are in a scenario where their jobs are being automated, this pressures against wage adjustments. If wages don't rise but inflation does, it means that people's purchasing power will diminish, and consumption will decelerate further.

In other words, if no one can afford anything because everyone's unemployed, who will support the economy?

On the top of THAT, we have COVID (which has not totally gone away) and war.
The hypothesis of high unemployment does not seem to be validated, at least at the moment?

Rather, the developed countries have had low birth rates for decades. The US was substantially ahead but I believe it has faded back in recent years.
While unemployment rates in the US are at record lows, the rest of the world is not experiencing the same fate. Eventually, if the world as a whole slows down, it's very likely this will pressure the US too.
e
As for automation, it is being seen as a response to an aging population. But the core problem with automation is that its main goal is to do more with less. At first, this only eliminated low-paying jobs; but now, with neural technologies, we have AIs that can write, paint, draw, and even translate audio (Google is releasing such a feature today).

In order for unemployment to rise, machines don't have to do everything better than us. All it takes is they do most jobs more efficiently and cheaply than we do. Our economy was designed by people, for people.

What happens when robots can produce all our wealth for "free", but no one can afford it? This is a question without a good answer. In such a system, our current capitalistic system would have to fall. It wouldn't necessarily give room to socialism, but what we could see is a system that is artificially stimulated.

We saw a taste of that artificially stimulated system, if only for different reasons, when COVID struck. But if eventually there are no jobs, this may be a new normal.

The challenge with the next decades is that this automation won't be gone; it'll be evermore efficient.

On the short and medium term, we are reaping the consequences of cutting costs: because companies are always trying to reduce wages and moving their productions to other countries, not only industrial production is concentrated at a few countries, but this also threatens specialized labor. Why would one invest on specializing if their profession is seen more and more like a commodity? Why study for several years and acquire a huge debt in education if companies are willing to pay less and less for it?

This is one of the reason some fields lack qualified people.
I think even perfect automation still creates a huge number of jobs for people looking to improve machines and tell them what to do.
Here's where you are wrong. I'll give you an example with chess.

Back when engines started to become great, but weren't quite better than humans yet, there was such a thing called a "centaur", which is a chess engine and a human playing together. They would complement each other.

As engines got better and better, centaurs started to have worse performance compared to engines alone.

Why? Because humans not only have a limited calculation ability, but also because they have biases towards certain playing styles that weaken them.
Computers have no such biases humans do, period.

As automation gets closer and closer to perfect, we will find out humans will slow machines down, leading to worse performance.
In some fields, machines almost don't need to "tell them what to do" anymore. GPT-3 can code on its own, and all it relies on is a human to tell, in plain English, e.g, "code a game for me with a ship dodging from asteroids".

And guess what: it will do just that.

Also on the way are systems that are "self-healing", i.e, they can repair themselves. No need for human operators to fix them.
For example: https://www.analyticssteps.com/blogs/se ... n-robotics
Valuethinker
Posts: 44460
Joined: Fri May 11, 2007 11:07 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by Valuethinker »

Booglie wrote: Thu May 12, 2022 11:43 am
Valuethinker wrote: Thu May 12, 2022 9:16 am
Booglie wrote: Wed May 11, 2022 5:03 am If anything, today has the potential to be worse due to an aging population and world economy deceleration.

Also, weak unionizing works both ways. Because workers are in a scenario where their jobs are being automated, this pressures against wage adjustments. If wages don't rise but inflation does, it means that people's purchasing power will diminish, and consumption will decelerate further.

In other words, if no one can afford anything because everyone's unemployed, who will support the economy?

On the top of THAT, we have COVID (which has not totally gone away) and war.
The hypothesis of high unemployment does not seem to be validated, at least at the moment?

Rather, the developed countries have had low birth rates for decades. The US was substantially ahead but I believe it has faded back in recent years.
While unemployment rates in the US are at record lows, the rest of the world is not experiencing the same fate. Eventually, if the world as a whole slows down, it's very likely this will pressure the US too.
Recoveries are more muted in some parts of the world - Europe in particular. Partly that's just timing - when the Covid lockdowns ended, how bad they were, & how much money the government was prepared to distribute into the economy. Those economies will in time catch up - the new news is the Ukraine war, which will have unpredictable effects - bad for European energy (and thus heavy industry & consumer spending), good for armaments makers, etc. If several million Ukrainians are going to be long term resident in western states, that takes them from a low GDP per capita economy into more prosperous ones - which will improve GDP overall.

The US has an unusually insulated economy when it comes to the Rest of the World. Or rather, it matters what the ROW sells to the US: the US imports a lot of raw materials & manufactured goods. Exports matter less to the US economy -- that's substantially a function of size (in the way the volume of a balloon grows faster than the surface area). Germany (not taking into account the rest of Eurozone) and Japan are anomalously large exporters.

So the US (and has, in the past) grown faster than the world as a whole.

China it is trickier. The size of China's boom since 2008 has made it critical in a number of areas in terms of demand for raw materials, as well as capital goods, luxury goods etc. And a Chinese slowdown which is around Covid hits supply chains all over the world.
As for automation, it is being seen as a response to an aging population. But the core problem with automation is that its main goal is to do more with less. At first, this only eliminated low-paying jobs; but now, with neural technologies, we have AIs that can write, paint, draw, and even translate audio (Google is releasing such a feature today).

In order for unemployment to rise, machines don't have to do everything better than us. All it takes is they do most jobs more efficiently and cheaply than we do. Our economy was designed by people, for people.
Someone still has to clean the toilets though, or look after us in our nursing homes. Their relative wages may rise compared to knowledge workers - but those jobs are still likely to be done by human beings (consider the earnings of a good plumber against your average university graduate, these days; electrical contracting costs have c. 2x since 2018 - ask me how I know :? ).
What happens when robots can produce all our wealth for "free", but no one can afford it? This is a question without a good answer. In such a system, our current capitalistic system would have to fall. It wouldn't necessarily give room to socialism, but what we could see is a system that is artificially stimulated.
If it is free, it will cost us nothing to buy it.

In practice if labour costs fall, over time so will the prices of the end products. There is a very famous paper (by William Nordhaus) of the cost of a Lumen of light v workers wages. With the LED lightbulb it has fallen by something on the order of 1/10,000 of what it was in 1600, say.
We saw a taste of that artificially stimulated system, if only for different reasons, when COVID struck. But if eventually there are no jobs, this may be a new normal.

The challenge with the next decades is that this automation won't be gone; it'll be evermore efficient.
I agree the dislocational effects could be large. As any industrial worker in the American "rustbelt" has known since the 1970s. If that happens sufficiently fast then you get the northern England scenario - the coal, steel and textiles goes, and nothing replaces them for a generation. They were crushed in the 1980s and it was into the 2000s before employment anywhere near full employment was reached. I can also see that if, for example, someone perfected self-driving cars (AVs) that would cost a lot of jobs (all those cab & uber drivers). Or if Amazon managed to fully automate its distribution centres - that's been a big growth area of employment.

However, and particularly in light of birth rates & lower levels of immigration, I suspect the problem for the employer of the future is finding workers. There are already fewer 18 year olds, and most western countries seem to be hardening against immigration -- regardless of who is in political office.
On the short and medium term, we are reaping the consequences of cutting costs: because companies are always trying to reduce wages and moving their productions to other countries, not only industrial production is concentrated at a few countries, but this also threatens specialized labor. Why would one invest on specializing if their profession is seen more and more like a commodity? Why study for several years and acquire a huge debt in education if companies are willing to pay less and less for it?

This is one of the reason some fields lack qualified people.
But that's things like truck drivers-- the big shortages. The problem is those jobs didn't pay well enough for the tough conditions that driving a long distance truck entail.

Large education debt burdens are primarily an American thing (the UK system of student loans is really a graduate tax, that cannot be called by that name).

University education has always been tangential to the job market. Lots of jobs now "require" a university degree but did not 50-60 years ago. The skills "taught" in university (I've spent enough time around universities to know that not everyone is imbued with them) are often not particularly applicable to the world of work.

I certainly know of several kids who have chosen to go the pilot route rather than the conventional university one - but that's because there is a shortage of airline pilots, as an older generation retires (and, post Cold War, there are fewer ex military pilots leaving the armed services).

I suspect the globalization of supply chains is going to pause, and even reverse. Covid, and now this war, have taught some ugly lessons regarding supply chain vulnerability. A war in Asia would do even more so.

I think companies are going to make even greater efforts to improve productivity because they will find it difficult to cut costs at the expense of their employees.
alfaspider
Posts: 4093
Joined: Wed Sep 09, 2015 4:44 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by alfaspider »

Booglie wrote: Thu May 12, 2022 1:55 pm
alfaspider wrote: Thu May 12, 2022 12:19 pm
Booglie wrote: Thu May 12, 2022 11:43 am
Valuethinker wrote: Thu May 12, 2022 9:16 am
Booglie wrote: Wed May 11, 2022 5:03 am If anything, today has the potential to be worse due to an aging population and world economy deceleration.

Also, weak unionizing works both ways. Because workers are in a scenario where their jobs are being automated, this pressures against wage adjustments. If wages don't rise but inflation does, it means that people's purchasing power will diminish, and consumption will decelerate further.

In other words, if no one can afford anything because everyone's unemployed, who will support the economy?

On the top of THAT, we have COVID (which has not totally gone away) and war.
The hypothesis of high unemployment does not seem to be validated, at least at the moment?

Rather, the developed countries have had low birth rates for decades. The US was substantially ahead but I believe it has faded back in recent years.
While unemployment rates in the US are at record lows, the rest of the world is not experiencing the same fate. Eventually, if the world as a whole slows down, it's very likely this will pressure the US too.
e
As for automation, it is being seen as a response to an aging population. But the core problem with automation is that its main goal is to do more with less. At first, this only eliminated low-paying jobs; but now, with neural technologies, we have AIs that can write, paint, draw, and even translate audio (Google is releasing such a feature today).

In order for unemployment to rise, machines don't have to do everything better than us. All it takes is they do most jobs more efficiently and cheaply than we do. Our economy was designed by people, for people.

What happens when robots can produce all our wealth for "free", but no one can afford it? This is a question without a good answer. In such a system, our current capitalistic system would have to fall. It wouldn't necessarily give room to socialism, but what we could see is a system that is artificially stimulated.

We saw a taste of that artificially stimulated system, if only for different reasons, when COVID struck. But if eventually there are no jobs, this may be a new normal.

The challenge with the next decades is that this automation won't be gone; it'll be evermore efficient.

On the short and medium term, we are reaping the consequences of cutting costs: because companies are always trying to reduce wages and moving their productions to other countries, not only industrial production is concentrated at a few countries, but this also threatens specialized labor. Why would one invest on specializing if their profession is seen more and more like a commodity? Why study for several years and acquire a huge debt in education if companies are willing to pay less and less for it?

This is one of the reason some fields lack qualified people.
I think even perfect automation still creates a huge number of jobs for people looking to improve machines and tell them what to do.
Here's where you are wrong. I'll give you an example with chess.

Back when engines started to become great, but weren't quite better than humans yet, there was such a thing called a "centaur", which is a chess engine and a human playing together. They would complement each other.

As engines got better and better, centaurs started to have worse performance compared to engines alone.

Why? Because humans not only have a limited calculation ability, but also because they have biases towards certain playing styles that weaken them.
Computers have no such biases humans do, period.

As automation gets closer and closer to perfect, we will find out humans will slow machines down, leading to worse performance.
In some fields, machines almost don't need to "tell them what to do" anymore. GPT-3 can code on its own, and all it relies on is a human to tell, in plain English, e.g, "code a game for me with a ship dodging from asteroids".

And guess what: it will do just that.

Also on the way are systems that are "self-healing", i.e, they can repair themselves. No need for human operators to fix them.
For example: https://www.analyticssteps.com/blogs/se ... n-robotics
In chess, there's a clearly defined goal and set of rules and a singular goal. The result is binary: win/lose. If we are talking about machines working in service of human masters, there's a very big question: what do we want them to actually do? As machines require less and less human input, that will simply result in the assignment of additional tasks, and more time thinking about what they will be assigned.

Imagine the worlds we could create if we actually had a omni-purpose robot that never needed repairs. You could essentially have real-life Minecraft where anything you can imagine can be created given sufficient resources. Yes, that makes the problem of basic existence relatively trivial, but it also greatly expands the universe of things we could seek to build.

Imagine if the denizens of what is now Dubai 200 years ago were told that someone would pitch their tents for them, drive their camels wherever they wanted, collect food, and even fish for them. The might have thought: wow, there's going to be nothing to do anymore! All my needs are provided. Now imagine the reaction you transported one of those Desert Nomads to the top of the Burj Khalifa in 2022. That's the equivalent of what an AI future you are describing could accomplish.

Of course there's always a Matrix-style dystopian options where the machines form desires of their own, but in that case the worry about human idleness similarly vanishes.
Booglie
Posts: 133
Joined: Sun Aug 15, 2021 2:17 am

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by Booglie »

Valuethinker wrote: Thu May 12, 2022 2:16 pm

In practice if labour costs fall, over time so will the prices of the end products. There is a very famous paper (by William Nordhaus) of the cost of a Lumen of light v workers wages. With the LED lightbulb it has fallen by something on the order of 1/10,000 of what it was in 1600, say.
Only electronics decrease in price as their efficiency increases. And even then, we have been reaching several limits. E.g:

[*] A computer from 10 years ago can do most basic tasks – browse the web, edit and print documents, do videoconferences. As their power increases, getting a new computer gets less and less justified. This is making the market more saturated. You can already see that in people taking longer to upgrade their devices.

[*] To add to that, companies are having trouble to increase power. Yes, Intel has been making more powerful processors, but at the cost of drawing more power. Apple has been able to revolutionize the market with their smaller processors, but they are expected to hit a roadblock when their processors reach 2 nm in size.

[*] Related to the two points below, an average user absolutely does not need the increase in processing power. To even justify that, you would have to improve on artificial intelligence assistants, but how essential are they, really? Most people I know don't even know their phones (yes, both Android phones and iPhones) come with a virtual assistant by default!

But I digress. The real point is that while technology is deflationary, most sectors are not. As you would expect, essential sectors prices (e.g, healthcare) have skyrocketed in comparison to TVs and technology.
Image



But that's things like truck drivers-- the big shortages. The problem is those jobs didn't pay well enough for the tough conditions that driving a long distance truck entail.
One more excuse for self-driving trucks: https://www.atbs.com/post/self-driving- ... ut-of-a-jo

Currently, they are only used in a few straight roads. But as the technology gets better and better, why justify human drivers at all, especially considering that AIs never get sleepy?

And again, driving AIs don't have to be perfect. All they have to do is to be much better than humans, which isn't as hard to achieve.

Large education debt burdens are primarily an American thing (the UK system of student loans is really a graduate tax, that cannot be called by that name).

I think companies are going to make even greater efforts to improve productivity because they will find it difficult to cut costs at the expense of their employees.
Millenials are earning 20% less on average than their parents, even though they have studied more: https://www.cnbc.com/2019/11/05/millenn ... cated.html

So, even on developed countries, the effects on driving costs down can already be seen. In the US, this also translates into the shrinking of the middle class. Income generations has been more and more polarized.

And that's on developed countries. Imagine how things go on development countries, where workers may not benefit from regulations.
halfnine
Posts: 1869
Joined: Tue Dec 21, 2010 1:48 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by halfnine »

SimpleGift wrote: Thu May 12, 2022 10:37 am
halfnine wrote: Wed May 11, 2022 1:28 pm From 1968 to 1973 the price of gold (black market rates) likely accurately reflected the collapse of Bretton Woods. From 1973 to 1980 rate of change in gold prices are fairly well correlated with the rate of change in CPI. Based on any data I have seen there is nothing to indicate that the unwinding of Bretton Woods persisted beyond 1973 as a major driving force for the pricing of gold.
You could be right. Not sure there is any way to disentangle the unwinding of Bretton Woods from the impact of double-digit inflation on the price of gold during the 1970s. My comment comes from looking at the real (inflation-adjusted) price of gold by decade, since 1900:
On the scale of a century, gold has kept up with inflation — but the real return of gold during the decade of the 1970s does stand out as a bit abnormal.
Here is the data I have from the 1970s (haven't looked at or crosschecked in ages so could be subject to errors)

Image

One could certainly draw lots of different conclusions from the data. But one can definitely conclude the extreme spike at the end of the 1970s was not the unwinding of Bretton Woods. Of course there is a bigger story that is not entirely shown on the chart and that is that gold would essentially (OTOH) quadruple from Jan 1979 to Jan 1980. The first doubling happened from January to November (I believe) and then from November to January gold would double again. A substantial chunk of that spike would quickly revert within a few months but it would take (again OTOH) until about 1990 to reach the inflation adjusted price of gold from Jan 1979.
User avatar
Topic Author
SimpleGift
Posts: 4165
Joined: Tue Feb 08, 2011 3:45 pm

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by SimpleGift »

halfnine wrote: Fri May 13, 2022 8:35 am Here is the data I have from the 1970s...
Interesting. Gold prices did accelerate in line with the two inflation spikes in the 1970s — and remarkably so.

However, the takeaway shouldn't be that gold in the free-floating era since 1971 has been a reliable inflation hedge. Yes, it did spectacularly well in the 1970s, but has been an unreliable inflation hedge ever since (chart below):
Investors seeking an inflation hedge (beyond TIPS, which just protect themselves) might be better served with an allocation to commodities, energy stocks, or REITs — or a 30-year fixed rate mortgage.
Call_Me_Op
Posts: 8559
Joined: Mon Sep 07, 2009 2:57 pm
Location: Milky Way

Re: T'was the Worst of Times — Case Study of 1973-1985

Post by Call_Me_Op »

Marseille07 wrote: Tue May 10, 2022 5:16 pm Excellent post.

Personally I'm surprised by the poor performance by 100/0 here. People often say equities beat inflation but that doesn't appear to hold water here.
They always add "over the longer term."
Best regards, -Op | | "In the middle of difficulty lies opportunity." Einstein
Post Reply