Vanguard: Predicting with CAPE

[Seasonal]

We've found that the CAPE ratio has more predictive power if interest rates and inflation levels are factored in. After all, one might expect that today's low interest rate/low inflation environment would make investors more willing to pay a higher price for future corporate earnings. The result of our approach is a "fair-value" estimate for the stock market that takes into account current economic and market conditions. (You can find our methodology and calculations here: Improving U.S. Stock Return Forecasts: A "Fair-Value" CAPE Approach, published in the Winter 2018 issue of The Journal of Portfolio Management.)

The CAPE ratio is high but not alarmingly above our current estimated fair-value range for the S&P 500 Index

the returns of U.S. stocks over the next ten years are likely to be below their historical averages because valuations are currently elevated. The most likely range of outcomes, according to our latest projections, is annualized returns of 4%–6% over the coming decade.

https://advisors.vanguard.com/VGApp/iip ... gTrmStkRts

[nisiprius]

I wish Vanguard would quit pulling this stuff.

Davis starts out well enough, by saying

Forecasting stock market returns has proven a perilous pursuit. Our research (and that of others) has demonstrated that many commonly used stock forecasting metrics have a poor track record, even at long investment horizons.

I sure wish he had stopped right there, and just left out the word "However" and everything that.

[DonIce]

How are they defining "fair value" CAPE? How much would, for example, a 1% change in interest rates affect it?

[warowits]

the returns of U.S. stocks over the next ten years are likely to be below their historical averages because valuations are currently elevated. The most likely range of outcomes, according to our latest projections, is annualized returns of 4%–6% over the coming decade.

Is this before or after inflation? The article says many times they are taking current inflation projections into account, but doesn't really say if their ultimate conclusion is nominal or real numbers. If it is 4-6% real then great, thats about what I figured.

[azanon]

Before "we" hammer market forecasting too hard, let's be reminded doing this sort of thing was one of Jack Bogle's fondest pastimes. Granted he didn't use CAPE.

[2015]

Before "we" hammer market forecasting too hard, let's be reminded doing this sort of thing was one of Jack Bogle's fondest pastimes. Granted he didn't use CAPE.

So what? Using appeal to authority bias does not make something worthless worthwhile. Forecasting is a great way to lose money. It's also a great way for VG to continue flushing marketing disguised as "information".

[willthrill81]

the returns of U.S. stocks over the next ten years are likely to be below their historical averages because valuations are currently elevated. The most likely range of outcomes, according to our latest projections, is annualized returns of 4%–6% over the coming decade.

Is this before or after inflation? The article says many times they are taking current inflation projections into account, but doesn't really say if their ultimate conclusion is nominal or real numbers. If it is 4-6% real then great, thats about what I figured.

1/CAPE is currently 'predicting' real returns of a little over 3% over the next decade, which is very similar to the 'prediction' offered by the average investor allocation to stocks.

But who cares what's going to happen over the next decade in stocks? BHs don't recommend keeping short-term funds in stocks for good reason, so stocks are a long-term play. Years ago, Kitces noted that subsequent 30 year returns after CAPE being in the top historic quintile versus the lowest quintile only varied by about +/- 1%. Now a 1% difference compounded over 30 years is a big deal, but no one should be using even historical average returns for our own planning purposes, IMHO. I believe that anyone whose plan depends on more than 5% real returns from stocks over the next 30 years is skating on thin ice, regardless of what the current CAPE is.

[heyyou]

This method of predicting the future will work well until it doesn't, then the new data will be incorporated into a shiny new, better method. If there wasn't so much money to be made when correct, we would care far less about predictions, knowing their imprecision. Predicting is just a sophisticated way to appeal to our inherent greed.

Save all you can, then adapt your spending to your assets, which will be easier since you are already conditioned to spending less, while longevity shows that almost all of us will not live 30 years past age 65 (another prediction).

[azanon]

Before "we" hammer market forecasting too hard, let's be reminded doing this sort of thing was one of Jack Bogle's fondest pastimes. Granted he didn't use CAPE.

So what? Using appeal to authority bias does not make something worthless worthwhile. Forecasting is a great way to lose money. It's also a great way for VG to continue flushing marketing disguised as "information".

This is Boglehead forum, is the so what. Whether I personally endorse forecasting would actually be besides the point.

I just want to make sure everyone that disagrees with forecasting, knows that they disagree with Jack. That's all. I wasn't making an argument for it, based on that, so no logical fallacy.

[siriusblack]

I think this is perfectly valid ... It's extremely sensible to me that CAPE should be expected to be higher in a low-interest rate environment compared to a high-rate environment. And, it's perfectly reasonable to me to talk about "expected returns" for equities as part of that analysis.

The equity "risk premium" should be expected to be influenced by prevailing interest rates. Think about it this way: If you could earn 15% by buying TIPS or something similar, why would you take more risk buying stocks? You would demand much lower P/E ratios in order to justify investing in stocks vs. bonds. So, if you end up buying stocks, you're saying the expected return of those stocks is greater than 15%. In that case, are you making a prediction? I would say yes but it's not a guarantee ... that's why they call it the "risk" premium in stocks.

That's all Vanguard is really saying here. So, in other words ... if folks look at CAPE and say "Whoa, that's way too high, CAPE is at near-record levels, I'm staying away from stocks", Vanguard is basically saying, "Not so fast! You might be comparing apples and oranges if you're expecting CAPE to ever get back to 1980 levels while interest rates are near historic lows."

This is a "prediction" or a "forecast" only insomuch as Vanguard is comparing the equity risk premium to prevailing rates and then extrapolating the % return that is apparently "priced in" for stocks at current levels. If bogleheads reject entirely the idea of the equity risk premium and/or the idea of expected returns based on valuation of stocks, then I really don't even know how to argue with that (maybe someone smarter here can help). For me personally, I think it's quite valid.

What I think bogleheads forget sometimes is that stocks go up over time FOR A REASON. That reason is corporate earnings that are retained (vs. distributed) and which therefore are reinvested and make the company more valuable. Warren Buffett has famously said that over a short period of time, stock prices can do pretty much anything (so in that sense, you CAN'T predict the short-term future); but over a very long period of time stock prices should be expected to go up based on these retained earnings (so in that sense, you CAN predict the longer-term future). It's not a "perfect" prediction because (again) prices can be irrational for some periods of time (thus the risk), or global disasters can happen, etc. ... BUT it's the reason we are all investing in the stock market in the first place ... the expectation (prediction) of future growth based on earnings. The ratio of prices to earnings matters because of the lower-risk bond interest rate relationship.

So ... to say you can't "time the market" is VERY different from saying you can't forecast expected returns. Your forecast may be wrong (that's why there is risk), but the prices are based on forecasts ... that's how pricing works!

[Beliavsky]

I wish Vanguard would quit pulling this stuff.

Do you think past average market stock returns will always be the best predictor of future ones? If not, what is your model of future expected returns?

[Beliavsky]

the returns of U.S. stocks over the next ten years are likely to be below their historical averages because valuations are currently elevated. The most likely range of outcomes, according to our latest projections, is annualized returns of 4%–6% over the coming decade.

Is this before or after inflation? The article says many times they are taking current inflation projections into account, but doesn't really say if their ultimate conclusion is nominal or real numbers. If it is 4-6% real then great, thats about what I figured.

1/CAPE is currently 'predicting' real returns of a little over 3% over the next decade, which is very similar to the 'prediction' offered by the average investor allocation to stocks.

But who cares what's going to happen over the next decade in stocks?

Every rational investor does. Even if your time horizon is 30 years, if you forecast stock returns to be no higher than bond returns over the next 10 years, why take the risk of investing with them? You can move into stocks if/when their forecasted returns are higher. The question is not whether good 10-year forecasts would be valuable but whether forecasts good enough to invest on can be made.

[willthrill81]

the returns of U.S. stocks over the next ten years are likely to be below their historical averages because valuations are currently elevated. The most likely range of outcomes, according to our latest projections, is annualized returns of 4%–6% over the coming decade.

Is this before or after inflation? The article says many times they are taking current inflation projections into account, but doesn't really say if their ultimate conclusion is nominal or real numbers. If it is 4-6% real then great, thats about what I figured.

1/CAPE is currently 'predicting' real returns of a little over 3% over the next decade, which is very similar to the 'prediction' offered by the average investor allocation to stocks.

But who cares what's going to happen over the next decade in stocks?

Every rational investor does. Even if your time horizon is 30 years, if you forecast stock returns to be no higher than bond returns over the next 10 years, why take the risk of investing with them? You can move into stocks if/when their forecasted returns are higher. The question is not whether good 10-year forecasts would be valuable but whether forecasts good enough to invest on can be made.

Fair enough. But CAPE has not shown itself to be a good market timing indicator, so the answer to the question you pose is no.

[nisiprius]

I wish Vanguard would quit pulling this stuff.

Do you think past average market stock returns will always be the best predictor of future ones? If not, what is your model of future expected returns?

I don't have a model. I don't believe they are predictable enough to "model." It is like asking what is my "model" for what team will win the Super Bowl over the next thirty years.

I read The Misbehavior of Markets: A Fractal View of Financial Turbulence by Mandelbrot and Hudson. It made sense, it was well reasoned, and rich in data. Ever since then, I've been waiting for someone to take that work forward and refine it--or refute it. Nothing. As nearly as I can tell, the financial community has decided that it is basically unanswerable. That it is true, but does not lead to any way for the investing community to make money, and that their best strategy is to ignore it and hope people will forget about it.

If you press me, I am willing to say that for planning purposes, over the next three decades, I expect stocks to return more than bonds, and bonds to return more than zero real, based on past averages, conventional wisdom--but only as a planning number, not a prediction.

[vineviz]

If you press me, I am willing to say that for planning purposes, over the next three decades, I expect stocks to return more than bonds, and bonds to return more than zero real, based on past averages, conventional wisdom--but only as a planning number, not a prediction.

This is your model, then.

And it’s probably even less predictive than Vanguard’s model.

[rkhusky]

If you press me, I am willing to say that for planning purposes, over the next three decades, I expect stocks to return more than bonds, and bonds to return more than zero real, based on past averages, conventional wisdom--but only as a planning number, not a prediction.

+1

I believe in the positive correlation between higher risk and higher return of a diversified portfolio over a sufficiently long period of time: stocks > bonds > cash.

[nisiprius]

Before "we" hammer market forecasting too hard, let's be reminded doing this sort of thing was one of Jack Bogle's fondest pastimes. Granted he didn't use CAPE.

So what? Using appeal to authority bias does not make something worthless worthwhile. Forecasting is a great way to lose money. It's also a great way for VG to continue flushing marketing disguised as "information".

This is Boglehead forum, is the so what. Whether I personally endorse forecasting would actually be besides the point. I just want to make sure everyone that disagrees with forecasting, knows that they disagree with Jack. That's all. I wasn't making an argument for it, based on that, so no logical fallacy.

Well, here's my appeal to Bogle. Since he was not doctrinaire, you can probably find a countervailing quotation, but I thought I should at least mention this one. It's longer than I want to quote in full, so there may be some selection bias in what I am choosing to quote.

How Important Is It to Forecast Future Returns?

There is no way for investors to avoid thinking about the future course of the financial markets. In this chapter, above all, I have tried to put into perspective the forces that drive market returns. They are worth knowing and understanding. But we must face the reality that, even if rational analysis of the relationship between investment fundamentals and speculation in investing gives us favorable odds (and no more than that) of accurate forecasting market returns, the game may not be worth the candle for the long-term investor....

[Bernstein and Arnott in an article in the Journal of Portfolio Management] concluded that "for most long-term investors, bull markets are not nearly as beneficial, and bear markets are not nearly as damaging as most investors seem to think..."

Irrespective of what the future holds, however, it seems to me that equities should remain the investment of choice for the long-term investor--the dominant component of a well-balanced asset allocation program.

So, invest with intelligence and common sense; engage in an enlightened and rational discourse when considering the future; always have some signficant portion of your assets both in stocks and in bonds; be sparing about precipitate and extreme changes in these proportions. And be skeptical about every prognostication you are given, including mine. If you have set an intelligent route toward capital accumulation, stay the course--no matter what.

[azanon]

Before "we" hammer market forecasting too hard, let's be reminded doing this sort of thing was one of Jack Bogle's fondest pastimes. Granted he didn't use CAPE.

So what? Using appeal to authority bias does not make something worthless worthwhile. Forecasting is a great way to lose money. It's also a great way for VG to continue flushing marketing disguised as "information".

This is Boglehead forum, is the so what. Whether I personally endorse forecasting would actually be besides the point. I just want to make sure everyone that disagrees with forecasting, knows that they disagree with Jack. That's all. I wasn't making an argument for it, based on that, so no logical fallacy.

Well, here's my appeal to Bogle. Since he was not doctrinaire, you can probably find a countervailing quotation, but I thought I should at least mention this one. It's longer than I want to quote in full, so there may be some selection bias in what I am choosing to quote.

How Important Is It to Forecast Future Returns?

There is no way for investors to avoid thinking about the future course of the financial markets. In this chapter, above all, I have tried to put into perspective the forces that drive market returns. They are worth knowing and understanding. But we must face the reality that, even if rational analysis of the relationship between investment fundamentals and speculation in investing gives us favorable odds (and no more than that) of accurate forecasting market returns, the game may not be worth the candle for the long-term investor....

[Bernstein and Arnott in an article in the Journal of Portfolio Management] concluded that "for most long-term investors, bull markets are not nearly as beneficial, and bear markets are not nearly as damaging as most investors seem to think..."

Irrespective of what the future holds, however, it seems to me that equities should remain the investment of choice for the long-term investor--the dominant component of a well-balanced asset allocation program.

So, invest with intelligence and common sense; engage in an enlightened and rational discourse when considering the future; always have some signficant portion of your assets both in stocks and in bonds; be sparing about precipitate and extreme changes in these proportions. And be skeptical about every prognostication you are given, including mine. If you have set an intelligent route toward capital accumulation, stay the course--no matter what.

For those that don't know (I'm would assume that you do Nisiprius), and to provide a clear basis for my original point about Jack Bogle, Jack was very fond of estimating future stock market returns, and in fact became quite well known for his formula for doing so which is: future market returns = Dividend Yield + Earnings Growth +/- Change in P/E Ratio. In the video interviews of Jack you can find around the net (such as Morningstar), in almost half of them, if not more, he'll use the formula to estimate returns. I presume that he did so, because he found it useful, and also presume that he wouldn't have done so had he considered it a waste of time.

[Seasonal]

I wish Vanguard would quit pulling this stuff.

Do you think past average market stock returns will always be the best predictor of future ones? If not, what is your model of future expected returns?

I don't have a model. I don't believe they are predictable enough to "model." It is like asking what is my "model" for what team will win the Super Bowl over the next thirty years.

I read The Misbehavior of Markets: A Fractal View of Financial Turbulence by Mandelbrot and Hudson. It made sense, it was well reasoned, and rich in data. Ever since then, I've been waiting for someone to take that work forward and refine it--or refute it. Nothing. As nearly as I can tell, the financial community has decided that it is basically unanswerable. That it is true, but does not lead to any way for the investing community to make money, and that their best strategy is to ignore it and hope people will forget about it.

If you press me, I am willing to say that for planning purposes, over the next three decades, I expect stocks to return more than bonds, and bonds to return more than zero real, based on past averages, conventional wisdom--but only as a planning number, not a prediction.

One use of a model for future returns is to decide how much to save each year. If you had a good idea of future returns you could calculate how much you'd need at retirement based on a guess about future spending and 4% (or 3%) SWR and then figure out how much to save to hit that target.

Another popular model is to save x% of income each year, but this makes implicit assumptions about future spending and returns (that may or may not be more reliable than would be generated by CAPE or some other model).

Absent a way of estimating future returns, how would you figure out how much to save each year?

I'm not disagreeing about the reliability of predictions, just wondering how you'd decide how much to save.

[Beliavsky]

I wish Vanguard would quit pulling this stuff.

Do you think past average market stock returns will always be the best predictor of future ones? If not, what is your model of future expected returns?

I don't have a model. I don't believe they are predictable enough to "model." It is like asking what is my "model" for what team will win the Super Bowl over the next thirty years.

I read The Misbehavior of Markets: A Fractal View of Financial Turbulence by Mandelbrot and Hudson. It made sense, it was well reasoned, and rich in data. Ever since then, I've been waiting for someone to take that work forward and refine it--or refute it. Nothing. As nearly as I can tell, the financial community has decided that it is basically unanswerable. That it is true, but does not lead to any way for the investing community to make money, and that their best strategy is to ignore it and hope people will forget about it.

If you press me, I am willing to say that for planning purposes, over the next three decades, I expect stocks to return more than bonds, and bonds to return more than zero real, based on past averages, conventional wisdom--but only as a planning number, not a prediction.

One use of a model for future returns is to decide how much to save each year. If you had a good idea of future returns you could calculate how much you'd need at retirement based on a guess about future spending and 4% (or 3%) SWR and then figure out how much to save to hit that target.

Another popular model is to save x% of income each year, but this makes implicit assumptions about future spending and returns (that may or may not be more reliable than would be generated by CAPE or some other model).

Absent a way of estimating future returns, how would you figure out how much to save each year?

I'm not disagreeing about the reliability of predictions, just wondering how you'd decide how much to save.

Yes, and how do you decide how much to allocate to stocks and bonds if you don't have an estimate for the excess return of stocks over bonds? Stocks are more volatile than bonds. Your allocation to stocks will be higher if your estimate of the equity risk premium is 6% rather than 1%. If you are a company or local government that needs to fund a pension plan, how do you set funding levels in the absence of forecasts of how your stock and bond investments will do?

[2015]

Before "we" hammer market forecasting too hard, let's be reminded doing this sort of thing was one of Jack Bogle's fondest pastimes. Granted he didn't use CAPE.

So what? Using appeal to authority bias does not make something worthless worthwhile. Forecasting is a great way to lose money. It's also a great way for VG to continue flushing marketing disguised as "information".

This is Boglehead forum, is the so what. Whether I personally endorse forecasting would actually be besides the point. I just want to make sure everyone that disagrees with forecasting, knows that they disagree with Jack. That's all. I wasn't making an argument for it, based on that, so no logical fallacy.

Well, here's my appeal to Bogle. Since he was not doctrinaire, you can probably find a countervailing quotation, but I thought I should at least mention this one. It's longer than I want to quote in full, so there may be some selection bias in what I am choosing to quote.

How Important Is It to Forecast Future Returns?

There is no way for investors to avoid thinking about the future course of the financial markets. In this chapter, above all, I have tried to put into perspective the forces that drive market returns. They are worth knowing and understanding. But we must face the reality that, even if rational analysis of the relationship between investment fundamentals and speculation in investing gives us favorable odds (and no more than that) of accurate forecasting market returns, the game may not be worth the candle for the long-term investor....

[Bernstein and Arnott in an article in the Journal of Portfolio Management] concluded that "for most long-term investors, bull markets are not nearly as beneficial, and bear markets are not nearly as damaging as most investors seem to think..."

Irrespective of what the future holds, however, it seems to me that equities should remain the investment of choice for the long-term investor--the dominant component of a well-balanced asset allocation program.

So, invest with intelligence and common sense; engage in an enlightened and rational discourse when considering the future; always have some signficant portion of your assets both in stocks and in bonds; be sparing about precipitate and extreme changes in these proportions. And be skeptical about every prognostication you are given, including mine. If you have set an intelligent route toward capital accumulation, stay the course--no matter what.

For those that don't know (I'm would assume that you do Nisiprius), and to provide a clear basis for my original point about Jack Bogle, Jack was very fond of estimating future stock market returns, and in fact became quite well known for his formula for doing so which is: future market returns = Dividend Yield + Earnings Growth +/- Change in P/E Ratio. In the video interviews of Jack you can find around the net (such as Morningstar), in almost half of them, if not more, he'll use the formula to estimate returns. I presume that he did so, because he found it useful, and also presume that he wouldn't have done so had he considered it a waste of time.

This is how these kinds of posts go awry. You use presumptions/extrapolations/projections regarding certain actions taken by Mr. Bogle and use those presumptions to imply that because Mr. Bogle took those actions the average investor who disagrees with those actions "disagrees with" Bogle. Again, who cares what Jack Bogle did? I repeat the assertion that the use of appeal to authority bias does not make a worthless action worthwhile.

I don't know which leg Mr. Bogle put his pants on first in the morning or why, but I do know that in complex adaptive systems such as investing the very best one can do is engage in a range of predictions. And even then there is no guarantee. The only usefulness of such an exercise is related to increased consideration given to risk mitigation.

If this is "disagreeing" with Mr. Bogle, then so be it. Like Charlie Munger, I do not respect professional boundaries.

[Seasonal]

I'm a bit surprised that those who disparage prediction models don't have an alternative for deciding how to save, allocate, etc. These are two of the most important things for financial planning and not having any answers or suggestions is not particularly helpful.

[GAAP]

But who cares what's going to happen over the next decade in stocks? BHs don't recommend keeping short-term funds in stocks for good reason, so stocks are a long-term play. Years ago, Kitces noted that subsequent 30 year returns after CAPE being in the top historic quintile versus the lowest quintile only varied by about +/- 1%. Now a 1% difference compounded over 30 years is a big deal, but no one should be using even historical average returns for our own planning purposes, IMHO. I believe that anyone whose plan depends on more than 5% real returns from stocks over the next 30 years is skating on thin ice, regardless of what the current CAPE is.

I "care" to the degree that it allows me to make a reasonable estimate of safe, variable withdrawal amounts from a retirement portfolio. I don't need "accuracy", just a general direction. CAPE basically gives a measure of the relative cost of equities -- and therefore the relative scale of returns and potential for price drops.

Note that I did not say "predictive" and that "relative" is just an view to where the market pricing appears to be.

My particular use case is for a portfolio that I am effectively managing as a near-perpetuity which must both provide income and support long term growth sufficient to maintain real value. This probably doesn't match the needs of most people in this forum.

[EnjoyIt]

Although I agree that we can't predict the future and the past does not necessarily have to repeat itself, I did enjoy the article. It can explain the higher valuations and why some of the low return predictions over the last 5 years have not been panning out.

[garlandwhizzer]

For what it's worth--granted, not much--I'll add my 2 cents to the essentially impossible task of predicting the future. The market clearly behaves more like quantum mechanics (no certainty, only probabilities) than Newtonian physics where future results of what is are accurately predictable given the state of current parameters. Having said that, for planning purposes we like to have predictions even if in the fullness of time they turn out to be wrong. Here's my guess: I think that given current parameters Vanguard's estimate of 4% - 6% which I assume to be nominal, nor real, is as good as anyones. There's clearly a wide range of outcomes clustered around those figures. I believe the most important point of all this forecasting chatter was made by willthrill81:

I believe that anyone whose plan depends on more than 5% real returns from stocks over the next 30 years is skating on thin ice, regardless of what the current CAPE is.

Garland Whizzer