For the last few years we have been poking fun (hopefully in a good-natured way) at the book Stocks for the Long Run.

About halfway through this historic equity bull market, in 1994 to be exact, Dr. Jeremy Siegel first published Stocks for the Long RunThe Washington Post called it “one of the ten best investment books of all time.”

Dr. Siegel posited stocks are less risky than bonds as holding periods lengthen. The following table displays the relative frequency of stocks outperforming either bonds or Treasury bills as a function of holding period. Specifically, stocks outperformed bonds over a thirty-year holding period 100% of the time from 1871 to 1993. From 1802 to 1993 stocks outperformed 97.2% of the time. The only time other than the present when stocks underperformed bonds over 30 years was the 1840s.

However, as the set of tables on the next page show, this trend has drastically changed in the wake of the financial crisis.

The first table shows returns through September 30, 2011. Bonds have completely dominated stocks over every tenor from one month to 30 years. Critics would say that September 30 is a favorable period for bonds as the all-time high in yields (low price) was September 30, 1981. This is true, but since the 1840s there have been numerous peaks in yields and none of those produced a bond outperformance over stocks as was seen since 1981.

So, to be fair, the second table below shows the most recent returns through May 31, 2012. Even when measured over this period, bonds still outperform stocks over most periods, only lagging at the 3-year and 30-year tenors.

According to Dr. Siegel’s book, this has not happened in hundreds of years and was never supposed to happen again.


We have called this generation’s outperformance of bonds over stocks the biggest investment theme everyone has gotten wrong. Over the years, when we note this, many say this is the most bullish argument they have ever heard for stocks, as this trend has to reverse. However, as the tables above highlight, this argument has been made for years and been wrong for years.

Add Dr. Siegel to this list.  His thesis was non-controversial in 1994 given the dogmatic belief returns increased with risk and stocks were a growth instrument while bonds were not.  Events interceded.  Now, in hindsight, we know that the biggest investment theme most missed over the last generation was the huge outperformance of bonds over stocks.

As the chart below shows, going against “one of the ten best investment books of all-time” has been a profitable idea.

It is really hard to identify a 160-year trend and write a book at the exact time it ends.  Does that make Stocks for the Long Run the worst-timed investment book of all-time?




Bonds for the Long Run  (August 9th, 2012)

Stocks Oil, Gold, Bonds for the Long Run? (January 17th, 2011)

Bonds Beat Stocks: 1981-2011 (October 31st, 2011)

Category: Data Analysis, Investing, Really, really bad calls

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.

38 Responses to “Revisiting Stocks For The Long Run”

  1. constantnormal says:

    I think that a lot of people are confusing nominal asset class performance and asset class performance when kept aloft by a fire hose of central bank liquidity … When the fire hose is turned off, the investing climate will change, and nobody can predict with any confidence how the changes will unfold.

    Perhaps Greenspan did usher in a Golden Age of bonds, but to me it looks more like a Golden Age of Bubbles.

  2. VennData says:

    ‘… as the tables above highlight, this argument has been made for years and been wrong for years…”

    Is NOT proof of anything, in fact lends legitimacy to the regression -to-the-mean argument used in financial markets. Since 2008, these guys have been telling you to eschew equities and you’ve missed the bull market of a generation… that little three year exception mentioned above… and that thirty year exception? That’s what Seigel’s talking about.

  3. machinehead says:

    ‘[Siegel's] thesis was non-controversial in 1994 given the dogmatic belief returns increased with risk and stocks were a growth instrument while bonds were not.’

    Eh, this is going a bit too far. That returns should increase with risk, and that stocks are growth instruments while bonds aren’t, aren’t merely ‘dogmatic beliefs.’

    Bonds, for instance, mature at no more and no less than their par price. They cannot offer principal growth. The apparent growth in the 30-year Treasury total return chart came from continuous rollovers as prevailing interest rates fell. Unless one believes we’re headed for persistent negative rates across the entire yield curve, the falling rate trend is just about OVER.

    As to the former assertion, bonds are now much more volatile than they were fifty or 100 years ago. High bond returns went hand in hand with higher risk. If that weren’t true, then the entire theory of finance since Harry Markowitz and William Sharpe would be useless.

    What Siegel got wrong was dismissing the impact — particularly the psychological impact — of deep drawdowns in stocks. Institutions with an indefinite life can take a detached view of a quarter-century S&P drawdown such as 1929-1954. But individuals cannot. Therefore, contrary to Siegel’s advice, people need multi-asset portfolios including bonds to moderate the extreme risk of stocks.

    Aside from this central error, Siegel’s book is one of the best histories of equities over the past 200 years. Siegel’s research was very thorough, and one can gain perspective from it even while dismissing Siegel’s conclusion.


    BR: Excepting that snafu with the faulty data. Other than that . . .

  4. machinehead says:

    By the way, the best history of bonds EVER is Sidney Homer’s magisterial A History of Interest Rates, which commences in Greco-Roman times.

    If you could read only one book about markets, this would be the one to choose.

  5. Orange14 says:

    Building on what constantnormal says above, how can bonds possible out perform equities over the next x number of years given the nominal low interest rates of treasuries right now. A lot of the gain in bonds has come from the dramatic lowering of interest rates by the Fed and we are now at pretty close to rock bottom for 30 year treasuries (can they go down to zero interest rates??). It’s always fun doing retrospective analyses to show how someone who wrote a book almost 20 years ago was wrong but it really doesn’t do much today (of course most of realized that the classic dumb investment book by Hassett and Glassman, “Dow 36,000) was wrong the day it was released).

    Remember, even a savvy bond investor such as Bill Gross got things terribly wrong in real time.

  6. Lee Adler says:

    If you ever catch Professor’s appearances on CNBC, he’s a good short term fade too. I have a pet name for him, but won’t use it here.

  7. Orange14 says:

    It’s always easy to poke fun retrospectively but much more difficult to get things right in real time. Look at the bad call Bill Gross made regarding Treasuries 18 months or so ago and he’s supposed to be one of the saviest bond investors around. Of course the classic dumb investment book in recent history was Hassett and Glassman’s ‘Dow 36,000′ which most of us just laughed off.

    I don’t think that it’s likely 30 year Treasuries will out perform equities over the next several years because of the recent Fed action. The chances for capital appreciation of the bond are slim to none and the rate of return on the coupon is just too low. If you look at the two curves in the chart you see two major periods when bonds appreciated much less than equities. It’s our challenge to figure out when those opportunities arise.

  8. chancypants says:

    Whenever I hear anything from Jeremy Siegel, I can’t help but hear this in the background. I like his tiny bicycle, too.

  9. streeteye says:

    The nature of a bond is you are going to get at best your principal and coupon back, in nominal terms. Unless, of course there is a default.

    Hard to make much of a case for bonds at 1.5%, when S&P dividend yield is 1.9%, plus the dividends on stocks are taxed at a lower rate.

    The scenario where bonds outperform is a deflation scenario…which is the last thing a heavily indebted society can afford, which would collapse the economy and other asset prices, and which the Fed is doing everything in its power to prevent.

    I don’t know if there is a bubble in gold or anything else, but there is definitely a bubble in government paper.

  10. bobnoxy says:

    The most amazing thing about Siegel is how he goes on t.v. and insists he’s been right all along! Anyone who followed his advice got hammered twice, and he smiles and says it’s a mirage. He ignores the survivorship bias that masks investor losses, not to mention how volatility compels real world investors to buy high and sell low.

    All things considered, one of the biggest hacks of our time, somehow never lacking for air time.

  11. yon’ QOTD seems rather apt, here..

    “In times of change learners inherit the earth; while the learned find themselves beautifully equipped to deal with a world that no longer exists.”
    -Eric Hoffer


    also, for my own Sense, “Eric” (as above) is no known Relation, of mine..

    though, iffin’ ya care for some real Intelligence, on matters of Nutrition (of the Physical kind..), from a Relative, of mine (not, that it should, really, matter..), see some of..

    I bring it up, primarily, because my Uncle, contra to Siegel(mentioned in the Post), never considered Himself ‘learned’, but, rather, “always Learning”..

    and, That, at the minimum, is one thing that I hope that He and I, do, share..~

  12. Jim Bianco says:

    Street Eye:

    The EXACT same argument was made in 1997 when 10-year Japanese JGBs broke below 2%. Since then the JGB has returned 49% (2.96% annualized) and the Nikkei has returned 29% (1.9% annualized).

    The real 99% on wallstreet are the bond bears are you do a good job of articualting their position. The problem is this position has been completed wrong for the years you have made it (all bond beras have been that way for years) and it was completely wrong in Japan.

    When it comes to markets, can something that 99% believe it be right? I say no which is why Siegel’s argument is not working.

  13. chris a says:

    No one is even mentioning the selective use of data to come up with his “history”. Barrons had done an analysis of the selective use of data that fit his thesis. It would be prudent to take a close look at the way his long run theses was created.

  14. chris a says:

    The Prof’s selective use of data should also be examined. Barrons had a piece about this not too long ago. He was very selective in the stocks he used, and it seems the data could have been manipulated to “prove” his thesis.

  15. lonr505 says:

    as mentioned the absolute biggest problem is data pre-1950-ish.

    The ideal measure of stocks would be to back-calculate the Russell 1000 back to 1812, but that’s nearly impossible as pre-1920-ish,, most stocks were thinly traded and share outstanding data is spotty.

    ANyone who says that they have a definite dataset/index re. the performance of stocks pre-1930 needs to put the entire dataset online for everyone to peer review.

  16. bobnoxy says:

    He ignores two big things that matter. Secular trends, and actual investor performance over time. How many people do you know, sitting fat and happy in retirement, that credit investing in stocks with having generated their wealth?

    Where are the customers’ yachts?

  17. idaman says:

    Street Eye is correct. The time frame you choose to compare investment returns will alter the results.

    Re-draw your chart again when the 10 year treasury is at 5%.

    If (as you *assume*) we are in for an extended bull market in bonds, that too will end, eventually.

    You may end up being right, but it is way to soon to draw such a conclusion.

  18. James P says:

    I wouldn’t ask why bonds are outperforming. I would ask why stocks are under performing. Read the Reformed Broker analysis posted earlier. Most management has been for the bottom line for a long time. It has resulted in the looting of companies. Even old family owned firms have the heirs just cashing in. No one wants to work. Everyone wants to live off rents. Where are the ETF’s or funds that own companies for long term strength and future cash flow? I don’t mean simply high dividend payers but also good growth management.

  19. Init4good says:

    also agree w streeteye, who is technically correct but….there are factors far outside the realm of the financial world which have enormous impact on the long term. For example, who knew that winning the last great war (wwII) would cause/ create a baby boom the likes of which has never before been seen in the human history? Did that “ppl bubble” have something to do with the demand for long-term dollars, starting roughly in 1978-early 1980′s to finance housing for these newly-minted families? Or that the demand itself would drive long term (mortgage) interest rates to 16%? Anectdote: My older brother had a 16% 30 yr mortgage on his first house. Or who could have predicted (besides the venerable Louis Rukeheyser) that long/short interest rates would fall for the next 18+ years? In retrospect it all seems so clear…
    So what else, of major significance, has happened to the population which greatly affects interest rates (the value of money)? I would say the number two issue (next to demographics) is the fact that it is no longer the “wealthy few” who see fit to own stocks and other securities. Fully 50% of the US population own stocks via mutual funds (at least truw 5 years ago). This in itself is cause for concern, and imo has a dragging effect on returns for stocks/ bonds. These investments are no longer “special enough” to be able to be selectively managed sufficiently to produce superior returns.
    Anything else? I say this current generation of youth having much less interest in owning cars, houses, and anything else considered a long-term investment – is a big deal. They are still entertained by the web, can go almost anywhere they want (virtually of course), can play games (or have sex) and communicate with people from all over the globe. They have shunned, (at least for now) the usual accoutrements of wealth. In that sense there is a lot that has changed.

  20. off_leash says:

    The analysis focuses on publicly traded equities. In the period since Professor Siegel’s book, there has been major growth in the private equity market and venture capital. Successful companies that previously would have gone public at an earlier stage now are fully priced by the time the general public gets to invest (just consider Facebook and Groupon). There’s historical merit to Siegel’s thesis but the world has moved on.

  21. Uchicagoman says:

    This is what the Coach-Potato portfolio is for….

  22. Uchicagoman says:


  23. [...] Is Jeremy Siegel’s Stocks for the Long Run the worst-timed book ever?  (Big Picture) [...]

  24. socaljoe says:

    “We have called this generation’s outperformance of bonds over stocks the biggest investment theme everyone has gotten wrong.”


    Every bond in existence had to be owned by someone.

  25. Lord says:

    I haven’t looked at this in detail but I expect this switch occurs whenever the market switches between secular bull and bear trends and it takes these to trigger reversals, but those trends rarely last more than 20 years. We look to the future, but all we can see is a mirror into the past.

  26. Ben says:

    I think your post is an important contrarian indicator. Mathematically, it will be very difficult for bonds from extremely low yield levels where they are today to outperform much of anything that goes up. Hence, it is like judging stocks vs. bonds at the end of 1999 – yes, if you judge from near the top of a bubble, conventional arguments will look stupid. Unfortunately, you’re looking at something extremely unsustainable.

  27. streeteye says:

    James – the difference is that the Japanese are big savers, and Americans are borrowers.

    In the US, if you have the the deflation that Japan experienced, all the mortgage and student loan borrowers, and more to the point the banks they owe money to, would be wiped out.

    Politically unlikely, but if someone wants a bond position as a hedge against a deflationary economic collapse I can see that.

  28. streeteye says:

    The only reasonable steady-state long term equilibrium for any asset class, if there was one, would be zero excess risk-adjusted return. If you think bonds are a great place at 1.5%, you’re basically saying every riskier asset class is going to have near 0 or negative returns, eg deflationary depression and economic decline. Possible, if we go Japan or worse, but not necessarily likely.

    There was a bubble in stocks, then a bubble in real estate, now a bubble in government paper. Each time some people said that asset class was the one that always worked over the long run.

  29. philipat says:

    In fairness, there were some who got it right. For instance, David Rosenberg, widely dismissed as an equity permabull, got it right about “Bonds have more fun”.


    BR: Rosie a Perma bull ?!?

  30. philipat says:

    Sorry, of course I meant to say “Permabear” not Permabull as written. Oh for an edit function!

  31. investingforaliving says:

    The use of the 30 yr treasury as being representative of ‘bonds’ seems like cherry picking to me. The bond market is a lot more than just the 30 year treasury. A more valid comparison would be to use the Barclays/Lehman Aggregate Bond Index, what all bond funds are measured against. I don’t have access to all the historical data but since Dec 1987, 25 years or so of history, the aggregate bond index has done something like 7.1% a year.

    Of course using a more representative index for the bond market weakens Mr Brianco’s argument somewhat. It is still a good argument but the argument is a lot more subtle. With this more valid comparison you need to start talking about risk adjusted returns and equity risk premiums. By this time most investor’s heads would have hit the keyboard….

  32. Jim Bianco says:


    This piece was a critique of Siegel’s book. He used the 30-year as that is the benchmark on Wall Street. The Lehman aggregate was invented in 1974 and cannot be used for long-term analysis. Sidney Homer used the same benchmark in his book. It is a valid benchmark for the bond market.

    Street Eye:

    Siegel’s book went back to 1802. In the last 210 years every conceivable scenario has happened … inflation, deflation, 15% long bonds and 2% long bonds. Yet his conclusion was from the 1830s to 1994 stocks outperformed bonds during every 30-year period measured. That is until he wrote the book!

    So, you don’t need to rationalize that this period is somehow different than anything seen in human history as this study almost cover all of human history. The fact is the long bond has made more money than the stock market and it is amazing how people absolutely refuse to admit this simple fact which is why it is the worst call of their career.

    Regarding your contention that bonds will be hard pressed to return more than 1.5% going forward, going forward is absolutely correct. But why do you think stocks will return more than 1.5% going forward? That was my Japanese example. 15 years ago the EXACT same argument s were made in Japanese and they proved just as wrong … in the last 15 years the 10-year JGB returned 3% as predicted (2% coupon and 1% capital gain as yields fell to 1%) but what shocked everyone was Japanese stocks returned less than 2%.

    The last 13 years (since Dec 31, 1999) the S&P 500 has returned and annualized 1.36% while the 30-year bond has returned an annualized 9.72% over the same period, the largest difference in favor of bond over such a period in 210 years of data. Other than Van Hoisignton (the greatest living money manager you never heard of) who got this trade right? Answer, no one regularly on CNBC (including me!)

  33. CANDollar says:

    What is the performance of stocks with a simple rules based risk management system such as going to cash when the 10 month SMA is breached to the downside (Fabers rule)?

    Similarly a simple portfolio hedge that caps drawdown at 10%-20% using 6 month puts would improve returns. It would cost in the order of 3% of equity performance on average during the period used and could be used only when technical and fundamental indicators show increased probability of a large drawdown. In the past 10 years there would have been only three periods where this would have been indicated.

  34. streeteye says:

    The main reason I think deflationary disaster is unlikely (although not impossible) is the Fed is scared to death of the possibility and determined to make sure it doesn’t happen.

    I think it’s bimodal, if there’s a muddle through and gradual return to inflation, then bonds return 1.5% nominal, 0 or negative after inflation, and equities have modest positive real returns.

    If the bond market is really and negative TIPS out to 20 years are the best asset class to be in, deflation is a horrible environment for stocks, no growth and no price increases, households stuck in increasingly underwater mortgages as prices and wages go down, more financial crises, defaults, political turmoil. If there’s financial crisis contagion and the Fed is out of ammo to prevent resulting demand shocks, it could happen. But not the way I’d bet.

  35. [...] for the long run. Bill Gross of PIMCO set up the debate, and others have piled in. The latest is this post from James Bianco reviewing Jeremy Siegel’s book, Stocks for the Long Run. Bianco points out that bonds have [...]

  36. CorvetteKid says:

    I respect Barry R., but his attacks on SFTLR and Jeremy Siegel are off base. We had a unique, literally once-in-a-lifetime BUBBLE in yields on risk-free assets, U.S. Treasury bonds. Once the yields crossed the double-digit threshold, there was no way that stocks would mop the floor with bonds on most measures going out 10-30 years.

    Second, if you held a diversified portfolio of DIVIDEND STOCKS over the last 5-10 years (or since the peak in 2000), you did very well. As an example, a simple mortgage REIT like Annaly Capital Management (NLY) has returned over 15% compounded since 1997 vs. 6% (approx) for the S&P 500.

    Third, I agree Van Hoisington is a great (bond) manager, but let’s see how he does with the 10-year at 1.65% and the 30-year at 2.75%. Is he investing his clients money for the next 5-10 years based on a Japanese Scenario with the 10-year and 30-year going to 1% and 2%, respectively ?

    No offense, and I am not an aggressive investor, but I don’t want to take that bet. I’ll take a diversified portfolio of dividend stocks, bond CEFs, and other income investments with a total portfolio yield of 6% or more. As long as I don’t lose 4% a year for a decade on the principal, I am likely to come out ahead.

    All dynasties come to an end — even the New York Yankees dynasty from 1927-1964.


    BR: You can disagree with my opinion, but the criticism of Siegel’s sloppy and incorrect data sources are from much tougher critics than I — its the WSJ and Biriyini.

  37. CorvetteKid says:

    Jim Bianco, you make some great points (I always read/see your commentaries) but just as U.S. treasury bonds were in a YIELD BUBBLE culminating in September 1981, Japanese stocks were in a bubble and even in the mid-1990′s were still expensive. The Japanese Nikkei 225 was yielding 0.7% in 1989 and P/E ratios were still over 40 in the mid-1990′s. I don’t think folks were ‘shocked’ that Japanese equities returned so little, though surprised might be an OK adjective.

    The U.S. has NOWHERE near the level of problems — financial, structural, valuation, or demographic — that the Japanese economy and stock market had to endure since 1989. Look at how little damage the NASDAW bubble did to our economy (unforunately, the housing bubble was worse). We got TARP done in a few weeks in September 2008. It took Japan years to get their ‘TARP’ with the unwinding of the keiretsu and other bank interlocking share holdings.

    The outperformance of risk-free assets like Treasury bonds the last 5, 10, and 15 years is a unique occurrence that is EXTREMELY UNLIKELY to persist going forward. It’s like someone calling HEADS for a coin flip 10 times in a row and being correct each time. Unlikely but theoretically possible. But I don’t think you or I would want to bet on HEADS another 10 times in a row, would we ?

  38. To quote Jan L. A. van de Snepscheut: “In theory, there is no difference between theory and practice. But, in practice, there is.”

    Any investing philosophy, theory or system that cannot adjust to a massive Fed intervention is flawed, or at least vulnerable to these sorts of events.