Thursday, March 24, 2011

100-Years of Inflation-Adjusted Stock Market History

When analyzing stock market performance over multiple years, it is important that you consider inflation-adjusted performance. The more years being evaluated, the more important it is to include the impact of inflation. If you don't adjust for inflation, you're comparing apples and bananas. Previously, I've suggested that you at least mentally factor in the impact; in this post, we'll update the 100-year stock market chart to make that "factoring in" explicit.

100-Year Stock Market Graph -- Inflation-Adjusted


100-year stock market history in  inflation-adjusted constant 2012 dollars

In the chart above (click to expand), the red line shows Dow Jones Index year-end closing prices in "then current" dollars -- as they were reported at the time. That is how you normally see them, and how they were displayed in my original 100-year stock market graph. In that original chart, you could see that the Dow is now more than 100 times higher than it was in 1900! But what does that really mean? Does it mean that if some (very) long-lived soul had invested $1 in the stock market in 1900 it would now be worth more than $100? Yes, but that's misleading because you're comparing apples and cumquats; the first amount is in dollars from 1900, the second amount is in more recent dollars. (Note: In this graph, closing prices are measured against the left vertical axis, which is log-based rather than linear. See About Stock Market Log Graphs for a discussion.)

Because of inflation, the purchasing power of a dollar was a lot greater in 1900 than it is now. That is, you could buy a lot more for $1 than you can now. For example, in 1900 a pound of butter cost $0.24; today I pay $4.49. On average, the purchasing power of a "1900 dollar" was more than 28 times that of a "2012 dollar." So, while the investor would have 100 times as much money, he would not, for example, be able to buy 100 times as many groceries.

Dow Jones Index Closing Prices in Constant Dollars

The blue line is inflation-adjusted; it takes that change in purchasing power into account. It does that by recalculating the closing prices and converting them all to dollars with the same purchasing power -- January, 2012 dollars. Now we see that in constant 2012 dollars the Dow has increased from around 2,080 in 1900 to 12,218 at year-end 2011. The increase is not only less than 100 times, it's less than 10 times.

Nominal vs. Real, Inflation-Adjusted, Stock Market Growth

The nominal compound annual return of the Dow from year-end 1900 to year-end 2011, excluding dividends, was 4.75%. That is, in then-current dollars, the price of the Dow increased on average 4.75% per year for more than 100 years. However, the real compound annual growth was only 1.6%. That is, when we use constant dollars, we discover that the purchasing power has only increased 1.6% per year. The difference between the two rates of return is inflation, which averaged about 3% per year. (Note: for reference, the tan dotted line at the bottom of the chart is yearly inflation, read off the vertical axis on the right side of the chart.)

The Long Flat Periods Revisited -- Now They're Worse!

In the original 100 Years of Stock Market History post we noticed that the stock market sometimes appears to get ahead of itself. Periodically, prices reach a peak that is not permanently surpassed until decades later. Unfortunately, viewed in constant dollars, those long flat periods become even longer.

In then-current dollars (the red line), the market closed at 969 in 1965, was relatively flat for about 17 years, and finally broke past 969 for good in 1982. In constant dollars (the blue line), it's a little scarier. In constant 2012 dollars, the 1965 close was 6909, and the real price was down (not "flat") for about 17 years. Instead of bottoming in 1974 (at 616 in then-current dollars), the market bottoms 7 years later in 1981. Finally, not until 1995 did the price permanently exceed the 1965 close. That long "flat" period was not 17 years, it was 30 years. (Picky readers will note that I'm making an assumption here....)

It gets worse. In then-current dollars, 1905 closed at 96. A 15 year flat period was followed by a huge, but temporary, spike that ended with the 1929 crash; the market finally passed 96 for good in 1933 -- 28 years after it first closed there. However, in constant dollars, the story is alarmingly different. In 2012 dollars, the market closed 1905 at around 2575. And, that 1981 bottom mentioned in the previous paragraph is at about 2103 -- even less than the 1905 close! It is not until 1983 that the market permanently closes above 2575. That's 78 years if you're counting.

Caveats/Notes

These results are based upon year-end data; results based upon daily or monthly data might be slightly less alarming. In addition, the CPI data is far from precise -- especially when you try to go all the way back to 1900 (See this article for a comprehensive discussion of some of the issues). However, I'm primarily trying to get the "big picture" -- I'm okay with some imprecision. Also, these results are based upon the DJIA (Dow Jones Industrial Average). However, the S&P 500 or any other broad-based index would yield similar results.

Finally, remember that this is just the price component of an investors' total return. Dividends are also an important component of total return. Well, at least they used to be....


Related Posts

100 Years of Inflation Rate History: an overview of inflation, and its impact on assets (stocks, bonds, housing), and The Declining Value of a Dollar.
U.S. Treasury Note Real Return History: Inflation-adjusted bond returns.
100 Years of Inflation-Adjusted Housing Price History: for comparison.
Inflation Calculator/Excel Spreadsheet: calculates inflation rate between any two years, and converts dollars to equivalent purchasing power.
100 Years of Stock Market History: the original post, with nominal prices only, and discussion of long flat periods.
The composition of long-term stock market returns: the contribution dividends have made historically.
About Nominal & Real Rates of Return: the difference between real & nominal returns.
For lists of other popular posts and an index of stock market posts, by subject area, see the sidebar to the left.
Copyright © 2011. Last modified: 3/6/2012

Share This Article

Delicious Bookmark this on Delicious To share via Facebook, Twitter, etc., see below.


Archives
100-year stock market history with real, inflation-adjusted closing prices

14 comments:

  1. Very interesting topic and well prepared. Thanks!

    ReplyDelete
  2. Thanks Anon. You're more than welcome.

    ReplyDelete
  3. You have done great work here.

    I've been trying to yell at the world that the hidden inflation tax has decimated the middle class and the poor.

    You used CPI here, but if you priced this in gas or food, returns would be even worse.

    Real Inflation and stagnant wages has killed the middle class.

    ReplyDelete
    Replies
    1. It is finally great to hear reason from
      these posts. You are absolutely right on
      the mark. The middle class is what made this country strong.Is that what big business
      wants? I don't mind paying more for things
      but for god's sake pay me enough to keep
      up and have some left over to retire on.
      I would rather me keep my SSN and invest it
      the way I want to. Call me stupid but why
      would I want to give my money away and live
      in poverty?

      Delete
  4. Thanks. Glad you're enjoying it.

    The gap between the highest income group and the lowest income group is indeed at record or near-record levels.

    I should point out, though, that these are actually not bad returns (don't forget, I've excluded dividends); I don't know of any asset class that performed better. It's just that inflation has historically been a major portion of what investors think of as their returns from stocks, housing and other long-term investments. As a result, when you remove the impact of inflation it's a bit of a shock.

    ReplyDelete
  5. Correct me if I’m wrong but I don’t think it was possible to buy the DOW index back then.
    Today we can purchase the ETF or some other instrument; I’ve yet to find a way to buy the index in 1900.

    Of course an investor could have purchased the 12 individual DOW components, but this complicates matters.

    All of the components (aside from GE) have gone bankrupt, were kicked out of the DJIA or they merged with other companies.
    If it was impossible to own the index, as components changed- an investor would have to rebalance their portfolios frequently.
    In other words; when stock market advocates praise the returns of the market with reinvested dividends, if one were able to hypothetically invest in 1900, and they tout today’s returns it is a bit of a lie.

    It would be very hard for a retail investor to sell and buy the DOW as it changed. Unless you were a trader and sold some stocks as they appreciated, to purchase new components as they were added it wouldn’t work.
    I was unable to find a way to purchase the DOW back then, perhaps it did exist???
    Otherwise, it’s a hypothetical and impossible scenario to use a buy and hold strategy that went on for a 100 year span.

    As far as real world investments that have real world returns, I can only think of two.

    Saving a 1 dollar gold coin from 1900 (purchased for face value) would be worth 90 bucks today.

    Don’t sell passbook savings short. We are used to today’s anemic savings rates, but it wasn’t unusual to get 5% and higher in the past.
    With compound interest and much higher interest rates, a dollar invested in 1900 would have returned better than a gold coin.

    ReplyDelete
  6. As I mention periodically, my stock market returns are indeed hypothetical -- theoretical. Broadly reported returns are almost always hypothetical; they ignore all kinds of real world realities including taxes and other expenses (partly because to do otherwise one would have to decide whose taxes, expenses, etc. to include). For what it's worth, the returns for bonds and housing are equally hypothetical. Personally, I think such numbers are still useful. However, in all cases, you need to realize that your personal returns are likely to be different from the theoretical returns.

    I do mention this periodically, but not in every single post. I've been thinking about writing a post to consolidate all of the caveats and qualifications in one place, and to discuss some of the implications. Your comment is a welcome reminder that such a post could be worthwhile. Thanks.

    ReplyDelete
  7. For more on how expenses can impact published/hypothetical returns see, Real World Expenses Reduce Published Market Returns.

    ReplyDelete
  8. Your chart does not include the dividend payouts, only the price index. Historically, dividends have provided about half of the return from stocks.

    ReplyDelete
  9. Hi - Good info on this site, but please can you make the full size graph images bigger and use less compression so they are easier to read?

    ReplyDelete
    Replies
    1. I'm a little confused by this since on my computer the images are quite legible -- even by these old eyes.

      Are you having difficulty even after you have clicked on the graphs to expand them? I'm not sure how to use less compression; can you say more?

      Delete
  10. Great blog thanks for all the information

    ReplyDelete
  11. Excellent Analysis! You might also mention the Survival Bias of the DJI30. I am sure you have that in mind, but never really say it. If you net out the effects of Survival Bias (re-indexing without the pain) the Real market return, net of dividends, is probably less than 0%. Better off to just buy Treasuries and hold for 100 years.

    ReplyDelete

No spam, please! Comment spam will not be published. See comment guidelines here.
Sorry, but I can no longer accept anonymous comments. They're 99% spam.