Over the last few weeks, we have discussed the questionable data and mediocre results of Jeremy Siegel’s Stocks for the Long Run (See this, this and this).

When we step back and take a look at The Really Long Run, we see a much clearer picture. The deep historical perspective as it pertains to the Dividend Yields on Equities vs Constant Yield on the 10 yr Bond may be revealing.

Equities have returned more from 1871 to 1956. An amazing 85 year outperformance run! After crossing over in ~1956, the 10 Year US Bond had a spectacular 56 year run of outperformance vs Equity Dividend Yield including all of the most recent bull runs.

Are Equities on the verge of another run of outperformance? (Note this is a very imprecise timing tool, measured in years and even decades, not nano seconds)

 

Equity Dividend Yield VS 10 Yr US Bond Constant Maturity Yield.
click for giant chart

Source: Global Financial Data

 

Thanks, Ralph !
Ralph Dillon rdillon@globalfinancialdata.com

Category: Dividends, Fixed Income/Interest Rates, Investing, Quantitative, Valuation

Please use the comments to demonstrate your own ignorance, unfamiliarity with empirical data and lack of respect for scientific knowledge. Be sure to create straw men and argue against things I have neither said nor implied. If you could repeat previously discredited memes or steer the conversation into irrelevant, off topic discussions, it would be appreciated. Lastly, kindly forgo all civility in your discourse . . . you are, after all, anonymous.

25 Responses to “140 years of Equity Yield vs US Bond Yield”

  1. dead hobo says:

    BR wondered:

    Are Equities on the verge of another run of outperformance? (Note this is a very imprecise timing tool, measured in years and even decades, not nano seconds)

    reply:
    ————-
    Hmmm … let me think. Will bonds do better than stocks or vice versa for the next 50 years or so, starting a few years from now, more or less? Good question.

  2. [...] Amazing chart – 140 years of equity dividend yield vs the 10-year US bond market.  Crossover happening?  (TBP) [...]

  3. wcvarones says:

    You are erroneously using the terms “outperformed” and “returned” when you are speaking strictly of yield while total return including price action was very different in many periods.

    ~~~

    BR: Fair point !

  4. rd says:

    Dividends were an important part of stock market returns until the late 90s. Since 1995, dividends have not come returned to the historic low dividend yield levels. This pairs well with Shiller’s CAPE which shows the same pattern of inordinately high PE valuations.

    It is clear for a while that Fed and other policies (including tax) are geared towards sustaining high equity valuations. Over the past few years, the Fed has also focused on pushing the yields of govenrment bonds down to unusually low levels as well.

    A jump to 3%-5% dividend yields would require, large sustained earnings increases, a massive re-allocation of cash from corporate coffers to investors, or a major whacking of stock prices. A return of CAPE valuations to norms would also require either a large sustained increase in earnings or a whacking of stock prices. Tobins Q value shows a similar pattern.

    Who knows what is actually coming over the next couple of decades but it is clear that financial instruments have left behind any semblance of historic sustainable fundamental norms over the past couple of decades leaving us in uncharted territoy (except for the technicians of course). The 2000 valuation peak seems to have been accepted by many people as just another market top instead of a massive mass hallucination on the scale of tulipmania. A new inflation adjusted peak higher than the 2000 peak is likely to be another decade or more away.

  5. constantnormal says:

    A more interesting chart would be a total return chart for bonds vs sticks … but that gets messy, as the results differ for different holding periods (what is the “maturity” of a stock?) … I suppose a good way to look at it would be a 3 dimensional surface, with date, gain/loss, and holding period as the three axes …

    And then we get into what constitutes an “average” stock or bond?

    Some questions do not lend themselves to being definitively answered.

  6. RW says:

    A chart of M-Square for 10-yr, 30-yr and S&P 100 and 500 would give you something closer to an apples-to-apples comparison I think; at least it would resolve the incommensurability problem WRT duration/maturity.

  7. Greg0658 says:

    we are here in this dual spectrum world and likely not going back .. but *:

    1> corporate stocks once bought are entrusted to the corporate board and their will
    (good/bad for you personally or your nation)
    a. commonman startups more difficult unless it supports a need inside the whales world
    b. board pay & expenses packages
    c. money flow direction
    d. once bought are at the markets mercy .. they are just another fiat currency

    2> the stock world is policed by the whole general public with tax dollars
    a. detail details details prove it in a court of law
    b. instruments are interconnected to the point the originators loose the chain flowchart

    3> bonds in their purist nature – are paid or bankrupted against .. hense mostly self policed .. whales grow or are mortally wounded

    * but not where I think we should be going in our highly evolved world in many areas BUT Political Science .. capitalism is an invasive species – we’ve got to the saturation point Again – interesting times indeed

  8. machinehead says:

    Don’t kid yourself: long run, there HAS to be an equity premium, because (despite what Jeremy Siegel claims) equities are riskier. Simple volatility shows this: average stock volatility of 16 percent is about twice that of bonds.

    But the equity premium pertains to total return, which can be received either in dividends or capital appreciation. Dividend policies were liberal up till the mid 20th century. Now dividend payout rates are low, and stock buybacks represent an alternative to dividends.

    Echoing what constantnormal said, dividend yield represents less than half of equity total returns, while coupon yield constitutes the lion’s share of bond returns. Thus the chart is an apple to half-an-orange comparison.

    Also, don’t take those 19th century Treasury yields literally. Before WW I, U.S. Treasury issues were few and far between, with no regular issuance calendar. There was no such thing as today’s well-populated yield curve. Furthermore, Treasury yields were distorted by the circulation privilege, such that most Treasuries were hoarded by banks at non-market yields.

    Railroad and municipal bonds were the prevailing reference for high-quality bond yields from 1871 to 1918. Full particulars are available in Sidney Homer’s History of Interest Rates.

    The notion of a 10-year constant maturity T-note yield in the 19th century is a lovely dream, but it is based on some heroic, desperate interpolations of extremely skimpy data. Arm yourself with a barrel of NaCl before using it. Or do it right and substitute New England municipal yields, which go back to the mid 19th century.

  9. Greg0658 says:

    I missed a major point:
    Interest Rate Growth on that old world pure Savings Account
    a. for the unconnected to the FIRE world
    b. for the busy working laborer that does not have a specialist for every need on the deductable payroll
    c. parents busy raise’g a family that are in the same situation

  10. wally says:

    The chart includes a period of about 100 years that corresponds to the growth of the American empire. Perhaps in a more mature developed world such an event will not occur again.
    That sort of macro background may be more important than any nuances found in the numbers.

  11. Vivian Darkbloom says:

    wcvarone has pointed out the obvious difference between yield and return. I’m glad Ritholz promptly acknowledged that. If capital appreciation would be included in stocks, the total return would heavily favor equities over the same period.

    Here’s another difference that may not be so obvious. I suspect the methodology used to make this comparison was to compare the yield on *newly issued* 10-year Treasuries at various points in time rather than the yield on all outstanding Treasuries. I doubt it is feasible to calculate the average yield on all outstanding Treasuries going back that far, so I strongly suspect that method was not used. Maybe this works itself out over time—it depends, I think, on the amount of outstanding Treasury debt at various points along the time line. For example, the spike in yields on *newly issued* Treasuries in the early 1980′s was due to the very high inflation during that period and the higher rates demanded by the public on newly issued debt. But that high rate would not be reflective of all outstanding debt at that point in time. Obviously, the reverse would be true at some time after rates start to fall–yield on newly issued Treasuries would tend to be lower than the average of all outstanding Treasuries.

    On the other hand, my guess is that the yield on equities was determined by taking the dividend yield on a broad base index and therefore it would reflect all outstanding equities. The comparison is therefore not perfect, even on a yield only basis.

    The other issue with respect to the fact that in more recent history bond yields have exceeded equity yields on dividends is that the stock market has shown, I think, a very marked increase in new listings say, since 1980. These new listings are typically over young, growing companies that are not in a position to pay dividends, or that prefer the growth opportunities of re-investing earnings over distributing them. This would skew the historical yield on equities if we are talking about, say, a total market index.

    Again, all of this is to say that yield is not everything and that even a yield comparison is not as straightforward as I suspect this chart leads one to think.

  12. Vivian Darkbloom says:

    Also, it would be interesting to see a comparison of the *after tax* yield (that’s what I care about). For the period 1913 through 1953, dividends were exempt from tax in all but the years 1936-1939. Small exemptions were granted from 1953 to 1985. From 2003 until now, the rate of tax on dividends has been significantly lower than on interest.

  13. znmeb says:

    This reminds me of Kenneth Fisher’s marvelous book of long-term historical charts, “The Wall Street Waltz”. Fisher came to the conclusion that common stocks outperformed all other forms of investment, though.

  14. rd says:

    vivian darkbloom

    The dividend yield shown is the S&P500 dividend yield, so these are established companies with proven track records and are replaced as they run into trouble, so the issue of start-ups etc is moot.

    Typical PE ratios of 10-25 with typical dividend yields of less than 6% imply holding times for stocks of a decade or more since you would not get your money back in less time than that unless the company’s hard asset liquidation value was a significant percentage of its stock price. So it is appropriate to be comparing equity dividend yields to 10 year or longer bonds instead of short-term bonds.

    FYI – the pre-1930s dividend yield and equity price charts are typically based on the Dow Jones Industrial average since the S&P 500 didn’t exist as an index before that, so there are issues with that part of the graph as well as the bond yields pointed out by machinehead.

    Stock buy-backs are nice, but shouldn’t change the PE or dividend yields, since the stock is still in competition with other income-producing instruments and the dividends and earnings should be increasing wit hthe stock price due to the decreased number of shares available.

  15. capitalistic says:

    We all know that studying historical charts from that far, is irrelevant. Bonds are more resistant to economic shocks (in my opinion). The reasons aren’t fundamental, but more behavioral. The same reason why people flock to real estate or Treasuries as a “safe asset’.

  16. philipat says:

    Bond Yields really took off after the ending of the Gold standard, probably not coincidental. Going forward, especially after the comps from a highly distorted present cycle (Including two bubbles and two busts) with Bonds starting from where they are now, the odds must surely be on Equities outperforming again on any reasonable risk/reward basis?

  17. Vivian Darkbloom says:

    rd

    Thanks for the response. Your statement that the yield was based on the S&P 500 was not reflected in the abstract of that study (or I missed it). Is your information based on the disclosure in the full study, or are you associated with the group that produced it?

    The rest of your comment is interesting, but obviously had little to do with mine. Thanks anyway.

  18. Greg0658 says:

    another point:
    as with US Social Security (I’ve heard) that 401K and corporate pension systems can be raided to pay bills (+ other endeavors) .. and replaced with IOUs / sometimes to the point of bankrupt worthless scripted promises

  19. rd says:

    vivian

    Top left of the graph says it is the S&P 500 Monthly Dividend Yield vs US 10 YR Constant MaturityBond.

    Just an engineer who looks for labeled graph titles and axes……

  20. socaljoe says:

    I wonder what the 10 year treasury yield would have been over the last couple decades without the FED and other central banks buying a few $trillion worth.

  21. [...] 140 years of equity yield vs US bond yield (Big Picture) [...]

  22. ElSid says:

    Timing tool? That’s America being full of itself — the blue line bottoms a few years after Bretton Woods — then comes the Nifty Fifty, Nixon blowing off the French, and Greenspan.

    Looks to me like nothing is going to return anything much at all, like Bill Gross was saying. It seems obvious to anyone who, oh, I don’t know, saw the .com bust coming or saw the housing bust coming.

    Nothing. Get over it. You grow crops too long it one field, they don’t grow no more, after a while.

  23. [...] Last week, we looked at 140 years of Equity Yield vs US Bond Yield. [...]

  24. [...] 140 years of Equity Yield vs US Bond Yield (September 4th, 2012) [...]