Thursday 14 June 2012

Severe Real S&P500 Bear Markets – June 2012 Update

As I write this post the S&P 500 is trading at 1,322.  If I correct for the devaluation of the US Dollar over the years (ie correct for inflation) we were at this value back in July 1997.  So in 15 years the value of companies in the S&P 500 have gone precisely nowhere.  Sure, as a private investor you would have been paid dividends over this period, but they have only averaged around 1.8% annually, making it pretty difficult to try and save for retirement.

My first chart today shows 3 periods in the US stock market since 1881 when similar conditions have prevailed.  I call these the historic severe bear markets and they are periods in time where from the stock market reaching a new high it then proceeded to lose in excess of  60% of its real (inflation adjusted) value.  The percentage change in value from the peak for each of these periods in time are shown in my second chart.  So what were these bear markets?

The first severe stock bear market (marked in purple on the chart) started with a new real (inflation adjusted) high being reached in September 1906.  Following this high the market then began declining with the period incorporating the 1907 Bankers Panic which was caused by banks retracting market liquidity and depositors losing confidence in the banks.  This occurred during a recession and there were a number of runs on banks and trust companies.  Additionally, many state and local banks went bankrupt.  From the high it took until January 1920 for the stock market to break through the 60% real loss barrier (a real loss of 60.9%) and then until December 1920 to reach its real low of -70.0%.  That’s 171 months a period of 14 years and 3 months.  The market then didn’t reach a new real high until September 1928, which is exactly 22 years, at which point it was rising at a rate of 25% year on year.  Unfortunately though, the market had less than 2 more years of positive real returns in it before hitting the next severe stock market.

This second severe stock bear (marked in blue on the chart) market started with a new real high being reached in September 1929.  This is the beginnings of the well known period of the Great Depression.  I won’t go into the history here as I’m sure it’s well known by all readers.  What is interesting however is that the US stock market passed through the -60% mark on a number of occasions.   From the high it took until January 1931 for the stock market to reach a real loss below the magic 60% mark at -62.0% and then until June 1932 to reach its real low of -80.6%.  The rate of decline was much faster than severe stock bear market one at 33 months.  However at this point the market never really recovered and dipped back below the real -60% mark in January 1933, July 1934, April 1938, June 1940, February 1941 and was back at -73.1% in May 1942.  That’s a period of 152 months or  12 years and 8 months.  Even 20 years later the market was still below the real -60% mark.  This bear market didn’t actually reach a new high until November 1958 which is a period of more than 29 years.  Again at this point the market was on a steep ascent rising at a rate of 27% year on year.

The third severe stock bear (marked in olive on the chart) market started with a new real high being reached in December 1968.  This period incorporated the stock market crash of 1973 to 1974 which came after the collapse of the Bretton Woods system and also incorporated the 1973 Oil Crisis.  So from the high it took until March 1982 for the stock market to reach a real loss of -60.9% and then until July 1982 to reach its real low of -62.6%.  That’s a period of 13 years and 7 months.  A new high wasn’t reached until December 1992 which is a period of 24 years.

So that brings me to today’s market which I show as the last line on my second chart, marked in red, showing the market we have been in since August 2000.  It started with the Dot Com Crash however we were unable to reach a new real high before the Global Financial Crisis took hold and we entered what is now being called the Great Recession.  I started discussing this period back in January 2010 and unfortunately things haven’t moved far since then.  At that time we were sitting at -38.1% and in 2.5 years we have only moved to -33.0%.  The good news is that I can’t yet call this period a severe stock bear market as we ‘only’ reached -58.6% in March 2009.  The bad news is that we are only 142 months into this one, which if history is any guide, leaves us plenty of time yet to see a severe stock bear market.  What will cause it?  There are many possibilities including Quantitative Easing leading to big inflation or Spain going bankrupt.  As always, I don’t claim to have any idea.  All I know is history suggests there is still plenty of time for it to occur.

As always DYOR.

Assumptions include:
-Inflation data from the Bureau of Labor Statistics.  June 2012 inflation is extrapolated.
-Prices are month averages except June 2012 which is mid market S&P 500 price on the 14 June 2012.


  1. Nice presentation (as usual). However, the thing missing here is the real return with dividends.

    For the UK (I assume S&P is similar), I think reinvested real returns have only been -ve in the decade 1911-20, when all other asset classes were too.

    In other words, even secular bear markets can be good times in which to invest relative to what else is on offer.

    1. Salis, You are a bit over-optimistic, loss lasting 10year periods(real, reinvested) are not that unusual. About 10% of all 10year periods are negative. If you only look at round number years, what about 2000-2010? That is negative (-23%), at least on the S&P500 it is, I don't have the FTSE data.

      That said, you are absolutely right that looking at the index data without dividends is pretty meaningless. The 29year bear market highlighted in the article becomes a 7year bear market when the dividends are factored in!

    2. @Paul, According to the Barclay Equity-Gilt study, the FTSE all share realised a total return of 0.6% for 2001-2010, the second poorest return since 1900, beaten only by -7.9% in 1911-1920.

      But yes, there was a bigger boom and bust in the US.

  2. Hi SG and Paul

    I agree that dividends are a critical feature. About a year ago I wrote this article highlighting the importance of dividends

    I think it's important to have a look at how dividend yields have behaved during these bear markets. The data is:
    - September 1906, start of the first bear, 3.8%
    - July 1918, equivalent point in first bear to today, 8.3%
    - September 1929, start of bear two, 3.0%
    - July 1941, bear two equivalent point, 6.8%
    - December 1968, start of bear three, 2.9%
    - October 1980. bear three equivalent point, 4.7%
    - August 2000, start of todays bear, 1.1%
    - Today, 2%

    The yields today are far lower than was achievable in the earlier three bears. To predict total return expected from equities, when forecasting retirement dates, I use Tim Hale's guidance from his great book which states an average 7.3% real return from equities. My US portion of my portfolio is going backwards from a capital perspective and achieving 2% dividends so I'm a long way behind.

    It's probably ok for me as I can just work for longer. I'm thinking retirees have it far worse. I'm sure plenty are using the 4% rule. If you have a large portion of US Equities and US Bonds you're probably feeling some pain about now.

    All that said, if I now consider a more balanced portfolio, I'm getting 4% divs in the UK and an even better 5.2% with my AUS equities which will compound nicely if reinvested.

    Of course, who knows what the future holds.


    1. Hi, I think that 7.3% return is a bit high. Jeremy Siegel "Stocks for the Long Run" produced 6.7%. I have personally reworked the superb data on Robert Shiller's "Irrational Exuberance" website and I came up with 6.3%.

      The dividends scene is interesting. Up until the 90's the US didvidend yield was nearly 5%. Since 1990 it has averaged barely 2%. This was due to the all the new tech companies, like Microsoft, that just did not pay dividends. What that means for total returns I am not sure.
      Cheers, Paul

  3. Hi Paul

    You might be right, 7.3% may be a bit bullish. I've just run a quick analysis on my dataset and get somewhere between the two. Using 1871 and 2012 as the dates I calculate a CAGR for the market at 4.1%, CAGR for inflation at 2.1% and an average of all monthly dividends at 4.5% for a total real annual return of 6.5%.

    How did you run the maths?

    Cheers, RIT

  4. Hi,

    Robert Shiller's website gives the prices, inflation and dividends and calculates the real values for monthly (US) data from 1870. I took the annual average figures (it reduces the data load and slightly smooths the data) and calculated the cumulative total annual returns by year. If you plot them on a log-linear plot the slope of the trend line gives the % total annual return. You can also see the true length and extent of bulls and bears. Very instructive. Cheers, Paul

  5. Hi Paul

    I might give that a try. I think I spot our difference though. When calculating the increase in inflation and the price I am calculating the Compound Annual Growth Rate.


  6. Sort of. It is probably that you are applying global values to them all, index, inflation and dividends and I am dealing with them on a year on year basis. For instance, in my case each years index is matched with that year's inflation and dividends. You are using the average inflation etc and applying them to all years. I wouldn't worry about it too much.

    Cheers, Paul