Valuation does not matter (may be)

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
long_gamma
Posts: 185
Joined: Mon Sep 14, 2015 4:13 pm

Valuation does not matter (may be)

Postby long_gamma » Tue Feb 14, 2017 6:47 am

Interesting presentation about US equity market sources of return. Nothing ground breaking, but have not seen presented this way.
https://blog.thinknewfound.com/2017/02/ ... et-follow/

Rolling 30 year sources of return for US market
Imagefree upload

Similarly for 3 years.
Imagefree photo hosting

Takeaway is for young investors valuation does not matter, just keep on investing regularly. But for older investor who is about to retire it matters a lot, especially since their asset base is larger and sequence of return matters just after they retire.

Wish they had done similar graphs in real terms.

They have a lot more good charts comparing recovery periods and other countries bull/bear markets here.
https://blog.thinknewfound.com/2017/02/ ... ll-market/
"Everyone has a plan 'till they get punched in the mouth." --Mike Tyson

dwickenh
Posts: 374
Joined: Sun Jan 04, 2015 9:45 pm
Location: Illinois

Re: Valuation does not matter (may be)

Postby dwickenh » Tue Feb 14, 2017 8:45 am

Thanks for the post long-gamma!!

I found the information may change some of the natural bias for looking at valuation.

Dan
The market is the most efficient mechanism anywhere in the world for transferring wealth from impatient people to patient people.” | — Warren Buffett

larryswedroe
Posts: 15404
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Valuation does not matter (may be)

Postby larryswedroe » Tue Feb 14, 2017 9:11 am

taking the wrong message from a chart.In fact valuations matter a great deal. Now that doesn't mean one should try to TIME the market. But valuations do impact the need to take risk.
Larry

User avatar
privatefarmer
Posts: 192
Joined: Mon Sep 08, 2014 2:45 pm

Re: Valuation does not matter (may be)

Postby privatefarmer » Tue Feb 14, 2017 9:49 am

the article says valuation changes certainly make up large part of near-term returns (~33%), and it says :

But over longer periods, the impact of valuations starts to approach zero as shorter-term fluctuations offset each other. Inflation and dividend yield together drive 70%+ of 30-year returns on average:
1.Dividends (44%)
2.Inflation (28%)
3.Earnings Growth (15%)
4.Valuation Changes (13%)


13% of the total return is not insignificant... its the difference of a 10% return vs an 8.7% return... Over decades that difference is huge. So I would say that, yes, valuations definitely matter.

User avatar
willthrill81
Posts: 742
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Valuation does not matter (may be)

Postby willthrill81 » Tue Feb 14, 2017 10:57 am

Valuations certainly matter, but not as much as some suggest.

"Notably, these results also suggest that even at valuation extremes, it would not be appropriate to adjust long-term equity returns by more than about 1% (or 100bps) in either direction. Which means even in today’s high valuation environment, where the Shiller CAPE is about 25.9, at the most it would only be appropriate to reduce long-term returns by 100bps."
https://www.kitces.com/blog/should-equi ... valuation/
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

long_gamma
Posts: 185
Joined: Mon Sep 14, 2015 4:13 pm

Re: Valuation does not matter (may be)

Postby long_gamma » Tue Feb 14, 2017 11:33 am

privatefarmer wrote:4.Valuation Changes (13%)


13% of the total return is not insignificant... its the difference of a 10% return vs an 8.7% return... Over decades that difference is huge. So I would say that, yes, valuations definitely matter.


One has to look at the counterfactual. Yes, Change in valuation accounts for 13% of the total return. But if one tries to time the market and sit out, then they may not participate 87% of the total returns.

US data does not give compelling argument for stating "Valuation matters" when one looks at rolling 30 year returns. But if one looks at Japanese market perspective, then probably can say valuations do matter.
"Everyone has a plan 'till they get punched in the mouth." --Mike Tyson

larryswedroe
Posts: 15404
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Valuation does not matter (may be)

Postby larryswedroe » Tue Feb 14, 2017 3:31 pm

will, fwiw, IMO that 1% figure is crazy wrong. Not even close to reality. And the people I talk to at high levels in the field would agree.
The ERP has historically ben 1%. I don't know anyone who thinks it is now 7% Stocks historically have returned about 7% real and most I talk to think somewhere between 4-5%, and that assume NO reversion to mean of valuations. If we do get that obviously returns will be much lower.
Wonder if the person giving this advice was saying in 2000 that should adjust returns no more than 1%? (Rhetorical)
Larry

User avatar
willthrill81
Posts: 742
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Valuation does not matter (may be)

Postby willthrill81 » Tue Feb 14, 2017 4:00 pm

larryswedroe wrote:will, fwiw, IMO that 1% figure is crazy wrong. Not even close to reality. And the people I talk to at high levels in the field would agree.
The ERP has historically ben 1%. I don't know anyone who thinks it is now 7% Stocks historically have returned about 7% real and most I talk to think somewhere between 4-5%, and that assume NO reversion to mean of valuations. If we do get that obviously returns will be much lower.
Wonder if the person giving this advice was saying in 2000 that should adjust returns no more than 1%? (Rhetorical)
Larry


Based on the data, that's not far off from what Kitces would argue based on the data, perhaps 5%. He argues that the real returns have historically been around 6.3%, so he's a more conservative than most from the outset.

From the sounds of it, you didn't read the article. Kitces' analysis is simple without being overly so and is quite compelling. Here is a very relevant section of it.

"For instance, the chart below shows the 30-year average annual compound growth rates for every starting year going back to 1871; time periods with starting valuations in the top 20% (a Shiller CAPE above 22) are colored red, while those with starting valuations in the bottom 20% (Shiller CAPE below 12) are green, and the rest (somewhere in the “muddled middle” 60%) are blue. Overall, the average 30-year real return is 6.3%, but while there is some variance of returns from high and low starting valuations, overall as expected the average of the high valuation zones is only 5.5%, while the lower valuation zones produced a 7.5% average return."

Image
https://www.kitces.com/blog/should-equi ... -valuation

Granted, if we broke this down further into something like deciles, this might be more relevant for us at our current CAPE values, but returns "much lower" than 4% do not seem to be supported by this data analysis at all. Perhaps you are referring to a shorter future time period?

Ironically, eight of the nine worst 30 year periods in terms of lowest CAGR were actually when the initial CAPE value was between 12-20, fairly close to the historic average.

Given that CAPE explains no more than 40% of the variance in market returns, and only that much 18 years out, this analysis seems to fit the data well to me at least. It's a moderately good metric for returns 15-30 years out but fairly poor outside that range.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

larryswedroe
Posts: 15404
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Valuation does not matter (may be)

Postby larryswedroe » Tue Feb 14, 2017 7:26 pm

Will
No I didn't read it and was going on what was posted. But here's the math.
The current cape 10 is 29, that inverts to a earnings yield of 3.4%. That's long way from 6.3%, not 1% difference.
Also note that the CAPE 10 has found to have 40% explanatory power at 10 years, not 18. And some Bogleheads did the work out to 20 years if memory serves and the correlation went higher, but longer it deteriorated.
Larry

User avatar
willthrill81
Posts: 742
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Valuation does not matter (may be)

Postby willthrill81 » Tue Feb 14, 2017 8:04 pm

larryswedroe wrote:Will
No I didn't read it and was going on what was posted. But here's the math.
The current cape 10 is 29, that inverts to a earnings yield of 3.4%. That's long way from 6.3%, not 1% difference.
Also note that the CAPE 10 has found to have 40% explanatory power at 10 years, not 18. And some Bogleheads did the work out to 20 years if memory serves and the correlation went higher, but longer it deteriorated.
Larry


Kitces addresses that in the article with the graph below. CAPE's correlation with nominal returns is indeed highest at about 8 years (R-square about 30%), but it's more strongly correlated with real returns at about 18 years (R-square almost 40%).

We're probably talking about different periods of time in which returns occur as well. Kitces' analysis was of 30 year CAGR, but it sounds like you're talking about a shorter period than that. I think that it's quite plausible that, based on the current CAPE, real returns will be closer to 4% ten years from now. But at that time frame, CAPE is only explaining around 30% of the variance, so the actual returns could obviously be significantly different.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

larryswedroe
Posts: 15404
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Valuation does not matter (may be)

Postby larryswedroe » Tue Feb 14, 2017 8:51 pm

Will of course they can be dramatically different, and the point is that the higher the CAPE not only does the potential dispersion of returns move to the left, the left tail increases because valuations can fall further.

Random Walker
Posts: 1418
Joined: Fri Feb 23, 2007 8:21 pm

Re: Valuation does not matter (may be)

Postby Random Walker » Tue Feb 14, 2017 11:12 pm

Larry,
When you say "valuations affect one's need to take risk", are you saying that high recent returns resulting in current high valuations result in decreasing one's need to take risk, and conversely that poor recent returns would result in low current valuations, causing one to need to take more risk in the future?

Dave

larryswedroe
Posts: 15404
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Valuation does not matter (may be)

Postby larryswedroe » Wed Feb 15, 2017 8:50 am

Random

Strong markets with rising valuations impact people differently based on their stage of investing. If you are late in your investment career bull markets are good, because they lower the need to take risk and you can take chips off the table, and if your marginal utility of wealth is now low you should do so. On other hand the worst thing for young investors in the early stages of their investment career is a bull market with rising valuations---like the 1990s. For them the need to take risk RISES because they have little investment capital at this stage--their largest asset is their labor capital--and now expected returns are now lower. So to achieve the same goal you need more equity risk. Not a good thing.

So in retirement or near it you root for bull markets and in accumulation stage, especially early, you root for bear markets (as long as that doesn't mean you lose your job!)

Hope that helps
Larry

asif408
Posts: 900
Joined: Sun Mar 02, 2014 8:34 am
Location: Florida

Re: Valuation does not matter (may be)

Postby asif408 » Wed Feb 15, 2017 9:28 am

larryswedroe wrote:Random

Strong markets with rising valuations impact people differently based on their stage of investing. If you are late in your investment career bull markets are good, because they lower the need to take risk and you can take chips off the table, and if your marginal utility of wealth is now low you should do so. On other hand the worst thing for young investors in the early stages of their investment career is a bull market with rising valuations---like the 1990s. For them the need to take risk RISES because they have little investment capital at this stage--their largest asset is their labor capital--and now expected returns are now lower. So to achieve the same goal you need more equity risk. Not a good thing.

Larry,

So would this mean that for younger investors there is even more reason for them to tilt their equities towards markets or segments of the market that have lower valuations (e.g., international, emerging markets)?

User avatar
willthrill81
Posts: 742
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Valuation does not matter (may be)

Postby willthrill81 » Wed Feb 15, 2017 10:01 am

asif408 wrote:Larry,

So would this mean that for younger investors there is even more reason for them to tilt their equities towards markets or segments of the market that have lower valuations (e.g., international, emerging markets)?


Not according to John Bogle.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
willthrill81
Posts: 742
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Valuation does not matter (may be)

Postby willthrill81 » Wed Feb 15, 2017 10:08 am

larryswedroe wrote:Will of course they can be dramatically different, and the point is that the higher the CAPE not only does the potential dispersion of returns move to the left, the left tail increases because valuations can fall further.


So Larry, do you think that GAAP data should be used to calculate CAPE or NIPA data instead, like Jeremy Siegel recommends? Siegel has found stronger correlations with market returns using NIPA data.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
nedsaid
Posts: 6642
Joined: Fri Nov 23, 2012 12:33 pm

Re: Valuation does not matter (may be)

Postby nedsaid » Wed Feb 15, 2017 10:15 am

larryswedroe wrote:taking the wrong message from a chart.In fact valuations matter a great deal. Now that doesn't mean one should try to TIME the market. But valuations do impact the need to take risk.
Larry


This is a good summary. Someone else raised a good point that valuations matter less to young investors than they do to retirees or near retirees.
A fool and his money are good for business.

larryswedroe
Posts: 15404
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Valuation does not matter (may be)

Postby larryswedroe » Wed Feb 15, 2017 11:01 am

will
yes, as there have been big accounting changes, which I noted, FA 142 and 144. There are other issues like much lower div payouts, which then can understate expected returns. I've written about this many times.
And Bogle has been wrong on two issue, one is value vs growth and that tell tale chart which doesn't show what he says it shows, and international diversification. I don't know a single financial economist who agrees. And IMO there is no logic to it, just home country bias.

Asif
The higher expected returns of int'l developed and EM are compensation for more risk. So that's a decision one has to make about risk taking. But if want or need higher returns that is certainly one way to get them. And I'd note the vast majority of the younger investors I would bet are VASTLY UNDERWEIGHTED as they "should" hold about 3/8 of their equity portfolio in developed markets and 1/8 in EM as that is about market cap. So they are betting against the collective wisdom of the market if they have less, or more.
Larry

asif408
Posts: 900
Joined: Sun Mar 02, 2014 8:34 am
Location: Florida

Re: Valuation does not matter (may be)

Postby asif408 » Wed Feb 15, 2017 12:14 pm

willthrill81 wrote:
asif408 wrote:Larry,

So would this mean that for younger investors there is even more reason for them to tilt their equities towards markets or segments of the market that have lower valuations (e.g., international, emerging markets)?


Not according to John Bogle.

No worries, I don't pray at the altar of St. Jack.

User avatar
grayfox
Posts: 4428
Joined: Sat Sep 15, 2007 4:30 am
Location: Earth

Re: Valuation does not matter (may be)

Postby grayfox » Wed Feb 15, 2017 12:49 pm

I have to comment here because this looks like a huge misunderstanding. First he writes valuation changes:

Over shorter time horizons, valuation changes start to dominate returns:

Valuation Changes (40%)
Dividends (25%)
Inflation (21%)
Earnings Growth (15%)


So far so good. Then he pulls the old switcheroo and writes valuations:

But over longer periods, the impact of valuations starts to approach zero as shorter-term fluctuations offset each other. Inflation and dividend yield together drive 70%+ of 30-year returns on average:

Dividends (44%)
Inflation (28%)
Earnings Growth (15%)
Valuation Changes (13%)


Nope. Now it is true that in the long term, valuation changes should approach zero and have no impact on long-term returns.

But that is different than saying that valuations have no impact. In fact, the effect of valuations has been greatest at about 20 years.
E.g. At higher valuations, initial dividend yield is lower, so the return from dividends is lower, even if no change in valuations.

:arrow: To summarize: Someone is confusing valuations with valuation changes. The first is a level, the second is the change in the level.
Gott mit uns.

User avatar
willthrill81
Posts: 742
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Valuation does not matter (may be)

Postby willthrill81 » Wed Feb 15, 2017 4:27 pm

larryswedroe wrote:will
yes, as there have been big accounting changes, which I noted, FA 142 and 144. There are other issues like much lower div payouts, which then can understate expected returns. I've written about this many times.


But when we use NIPA data as Siegel advocates, the market is not grossly overpriced as Schiller's use of GAAP data suggests, only 'slightly' overvalued.

Given that the NIPA data results in more predictive power than the GAAP data, combined with the sound logic for using it, this is partly why I'm dubious as to the statement that markets are far overvalued at present.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

hilink73
Posts: 47
Joined: Tue Sep 20, 2016 3:29 pm

Re: Valuation does not matter (may be)

Postby hilink73 » Wed Feb 15, 2017 4:47 pm

larryswedroe wrote:will

And Bogle has been wrong on two issue, one is value vs growth and that tell tale chart which doesn't show what he says it shows, and international diversification. I don't know a single financial economist who agrees. And IMO there is no logic to it, just home country bias.
Larry


I've read about the telltale chart every so often here.
Could somebody point me to what's that about?
Thanks.

larryswedroe
Posts: 15404
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Valuation does not matter (may be)

Postby larryswedroe » Wed Feb 15, 2017 5:11 pm

will
Certainly I NEVER said this market is vastly overvalued, just highly valued, even with the adjustments. But rationally high, just lower ERP

hilink
Short answer is Bogle made presentation/speech in which he showed chart, telltale chart, showing that value stocks and growth stocks had same returns and just went through cycles. The problem is that is chart doesn't show that. It shows the performance of ACTIVELY MANAGED growth and value funds, both of which style drift all over the place. All one need do to prove Bogle wrong is go to Ken French's website and look at returns. Amazing to me that someone as smart as Saint Jack would make that error, but worse, persist in telling that tale after he's showed it's wrong. And people keep referring to it even here after it's been explained why he is wrong and the telltale chart provides no such evidence
Larry

User avatar
willthrill81
Posts: 742
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Valuation does not matter (may be)

Postby willthrill81 » Wed Feb 15, 2017 5:12 pm

larryswedroe wrote:will
Certainly I NEVER said this market is vastly overvalued, just highly valued, even with the adjustments. But rationally high, just lower ERP


Gotcha. :)
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

larryswedroe
Posts: 15404
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Valuation does not matter (may be)

Postby larryswedroe » Wed Feb 15, 2017 6:39 pm


User avatar
willthrill81
Posts: 742
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Valuation does not matter (may be)

Postby willthrill81 » Wed Feb 15, 2017 7:02 pm



Thanks a lot for the link. That's a great piece and an argument that I've been thinking about myself for a while now, though not in the specific terms you laid out.

If we were able to quantify the 'adjustments' you mention in that article, it seems that might be very close to a 'fairly' valued market, or at least barely above where it 'should' be.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

larryswedroe
Posts: 15404
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Valuation does not matter (may be)

Postby larryswedroe » Wed Feb 15, 2017 7:33 pm

Will
Well fairly valued is fine, but even if you make adjustments of say 5 in the CAPE 10, that brings it to 24 now. Which leaves a real expected return of about 4%. Now versus tbills that is about 3.5% premium. That's well below historical average, but IMO rational given the changes I discussed. So not overvalued but highly valued
Hope that helps
larry

TJSI
Posts: 267
Joined: Thu Jan 27, 2011 4:03 pm

Re: Valuation does not matter (may be)

Postby TJSI » Wed Feb 15, 2017 9:09 pm

Note on the analysis:

This is not an analysis of the effect of valuation changes but rather the effect on initial valuation on subsequent return.

TJSI

lsp12
Posts: 18
Joined: Thu Dec 29, 2016 12:20 pm

Re: Valuation does not matter (may be)

Postby lsp12 » Thu Feb 16, 2017 9:59 am

One of my issues with Schiller and CAPE is that it doesn't account for changes in interest rates/discount rate (as far as I know).

The capital asset pricing model says that equity valuation is a function of the equity risk premium, beta (one for the overall market), and the risk free rate of return. So, as I look, for example, at the 30 year gov't bond rate, it has been a pretty straight down hill slope for the past 30 years. Now, of course, that decline reflects many things, including inflation expectations, but we all know there are many other factors that are independent of inflation. So, all things equal, the CAPE would go up when the risk-free rate of return goes down. So, if comparing the CAPE over time, IMO you cannot do that without considering the decline in underlying interest rates.

More intuitively, I think people recognize that, when alternative investments yield less (e.g. fixed income), then they will pay more for the expected future value of equities. You hear that all the time in the current low-interest rate environment.

None of this is to say that valuations don't matter -- of course they have to. But, if you compare valuations (or CAPE, as a proxy), you must account for the expected rate of return of our other options.

If you believe the equity risk premium is 7%, then if the risk free rate of return is 6% , the market discount rate for equities is 13%; if interest rates are 3%, then the discount rate is 9%. Compare the NPV of the same fixed future cash flow under those two scenarios -- PE of 1/.13 = 7.7 x in the first case, PE of 11.1x in the second. I.e. you've seen an expansion of the PE by 44%. Yet, they are both 'fairly valued,' with the same NPV due to the lower interest rate scenario in the second instance. Of course, the real world isn't so simple, but the underlying point remains valid.

TJSI
Posts: 267
Joined: Thu Jan 27, 2011 4:03 pm

Re: Valuation does not matter (may be)

Postby TJSI » Thu Feb 16, 2017 11:13 am

Note on Note:

After getting some sleep, I see that my note only adds confusion. I was referring to the to the Michael Kitces analysis not the article first quoted. Mr Kitces found that indeed starting valuations at a 15 year span had a very significant effect on returns. After 30 years, the effect was greatly reduced.

TJSI

User avatar
willthrill81
Posts: 742
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Valuation does not matter (may be)

Postby willthrill81 » Thu Feb 16, 2017 11:52 am

TJSI wrote:Note on Note:

After getting some sleep, I see that my note only adds confusion. I was referring to the to the Michael Kitces analysis not the article first quoted. Mr Kitces found that indeed starting valuations at a 15 year span had a very significant effect on returns. After 30 years, the effect was greatly reduced.

TJSI


He found that CAPE (using the inferior GAAP data by the way) predicted real (inflation adjusted) returns best 18 years into the future, at which point it explained roughly 40% of the variation, and then tailed off from there.

So you can look at that as a bit of a 'glass half full or half empty' way. Yes, it's a significant predictor of long'ish' term returns, but even then it's still not accounting for even half of what's going on.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

larryswedroe
Posts: 15404
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Valuation does not matter (may be)

Postby larryswedroe » Thu Feb 16, 2017 12:53 pm

Isp

the CAPE would go up when the risk-free rate of return goes down.


Unfortunately, this is one of the most common mistakes people make, believing this should be true. Economists call it the money illusion. It's just not necessarily correct. From a piece I wrote on the subject


The belief that nominal interest rates should matter to stock valuations is based on the idea that bonds are competing instruments for stocks. But those that use metrics like comparing bond yields to P/E ratios to determine “fair value” are making an error by thinking of things in isolation; thus, they fail to see the whole picture. In this case, the belief is based on what is called the “money illusion.” It’s one that has great potential for creating mistakes because one of the most popular indicators used by investors to determine if the market is under- or overvalued, is The Fed Model.

In 1997, in his monetary policy report to Congress, Fed Chairman Greenspan indicated that changes in the ratio of prices in the S&P 500 to consensus estimates of earnings over the coming 12 months have often been inversely related to changes in long-term Treasury yields. Following this report, Edward Yardeni, speculated the Fed was using a model to determine if the market was fairly valued — how attractive stocks were priced relative to bonds. The model, despite no acknowledgement by the Fed, became known as the Fed Model.

Since Yardeni coined the phrase, the Fed Model as a valuation tool has become “conventional wisdom.” Unfortunately, much of conventional investment wisdom is wrong. There are two major problems with the Fed Model. The first is that the expected return of stocks isn’t determined by their relative value to bonds. Instead, the expected real return is determined by the current dividend yield plus the expected real growth in dividends. To get the estimated nominal return you add estimated inflation. This is a critical point that seems to be lost on many investors. The result is investors who believe low interest rates justify a high valuation for stocks without the high valuation impacting expected returns are likely to be disappointed. When P/Es are high, expected returns are low, and vice versa, regardless of the level of interest rates.

The second problem with the Fed Model is it fails to consider that inflation impacts corporate earnings differently than it does the return on fixed income instruments. Over the long term, the nominal growth rate of corporate earnings has been in line with the nominal growth rate of the economy. Similarly, the real growth rate of corporate earnings has been in line with the real growth of the economy. Thus, in the long term the real growth rate of earnings is not impacted by inflation. On the other hand, the yield to maturity on a 10-year bond is a nominal return — to get the real return you must subtract inflation. The error of comparing a number that isn’t impacted by inflation to one that is leads to what is called the “money illusion.” Let’s see why.

To keep our example simple we’ll assume that there’s no risk premium in the yield on nominal bonds for unexpected inflation and there’s no liquidity premium in the yield on TIPS (Treasury Inflation-Protected Security). Let’s also assume the real yield on a 10-year TIPS is 2 percent (wouldn’t that be nice), and the expected long-term rate of inflation is 3 percent. Thus, the 10-year Treasury bond would be 5 percent and the fair value for stocks would be at a P/E of 20. Now lower our assumption of inflation of 3 percent to 2 percent. This would cause the yield on the 10-year bond to fall from 5 to 4 percent, causing the fair value P/E to rise to 25. But, since inflation doesn’t impact the real rate of return demanded by equity investors, it shouldn’t impact valuations. In addition, as stated above, over the long term there is a strong relationship between nominal earnings growth and inflation. In this case, a long-term expected inflation rate of 2 percent, instead of 3 percent, would be expected to lower the growth of nominal earnings by 1 percent, but have no impact on real earnings growth (the only kind that matters).
Because the real return on bonds is impacted by inflation while real earnings growth is not, the Fed Model compares a number that is impacted by inflation with a number that isn’t — causing the money illusion.

Now let’s consider what would happen if the real interest rate component of bond prices fell. The real rate is reflective of the economic demand for funds. Thus, it’s reflective of the rate of growth of the real economy. If the real rate falls due to a slower rate of economic growth, interest rates would fall, reflecting the reduced demand for funds. Using the earlier example, if the real rate on TIPS fell from 2 percent to 1 percent, that would have the same impact on nominal rates as a 1 percent fall in expected inflation, and, thus, the same impact on the fair value P/E ratio — causing fair value to rise. However, this too doesn’t make sense. A slower rate of real economic growth means a slower rate of real growth in corporate earnings. Thus, while the competition from lower interest rates is reduced, so will be future earnings.

Since corporate earnings have grown in line with nominal GNP growth, a 1 percent lower long-term rate of growth in GNP would lead to 1 percent lower expected growth in corporate earnings. The “benefit” of falling interest rates would be offset by the equivalent fall in future expected earnings. The reverse would be true if a stronger economy caused a rise in real interest rates. The negative effect of a higher rate of interest would be offset by a faster expected growth in earnings. The bottom line is there is no reason to believe stock valuations should change if the real return demanded by investors hasn’t changed.

When Clifford S. Asness studied the period 1881–2001, he concluded the Fed model had no predictive power in terms of absolute stock returns — the conventional wisdom was wrong. Asness also concluded that over 10-year horizons the E/P ratio does have strong forecasting powers. Thus, the lower the P/E ratio, the higher the expected returns to stocks, regardless of the level of interest rates, and vice versa.

There’s one other point to consider. A stronger economy, leading to higher real interest rates, should be expected to lead to a rise in corporate earnings. The stronger economy reduces the risks of equity investing. In turn, that could lead investors to accept a lower risk premium. Thus, it’s possible that higher interest rates, if caused by a stronger economy and not higher inflation, could actually justify higher valuations for stocks. The Fed Model, however, would suggest that higher interest rates means stocks are less attractive. The reverse would be true if a weaker economy led to lower real interest rates.

lsp12
Posts: 18
Joined: Thu Dec 29, 2016 12:20 pm

Re: Valuation does not matter (may be)

Postby lsp12 » Fri Feb 17, 2017 9:14 am

Larry:

Thank you for responding. Very interesting comments, though I don't necessarily agree with them. I've added comments in the text below.

First, here are a few (sincere) questions for you: do you believe that
1. the current valuation of stocks reflects, at least to a pretty significant extent, the market's assessment as to the NPV of future cash flows from those stocks? (not all the time, e.g. the Nasdaq bubble of 2000 was, IMO, based on a lot of assumptions about future growth in stock prices, as opposed to any coherent assessment of future cash flows, ie. a bubble predicated on the greater fool theory)
2. assuming the answer to #1 is yes, would, ceterus paribus, a decrease in the real rate of interest change the NPV of those cash flows?

A couple more questions:
3. can the real rate of interest every actually change, not due to changes in expected underlying economic growth, but for other reasons, e.g. a Fed pumping money for some reason
4. Is there a distinct possibility that the Fed's view of future economic growth (which can influence their interest rate decisions) can differ from the market's view of the future growth?

Obviously, any methodology of comparing p to (past) e has intellectual shortcomings, as the market is really all about the future. Moreover, trying to tease out the underlying rationale for changes in nominal interest rates (was it due to changing inflation expectations, was it due to changes in market driven supply/demand for money, was it due to Fed policy changes?) and how that relates to changed expectations about future corporate profitability, seems difficult at best.

So, I come back to an intellectual conclusion: all things being equal, if nominal interest rates fall, that is likely to increase the NPV of future cash flows from stock. Of course, that conclusion has an underlying assumption: that those future cash flows haven't been affected/reduced. And, of course, that is a huge assumption. However, it seems to me that basically saying changes in nominal interest rates have no effect on the NPV of future equity cash flows (because, if they do, they must affect stock prices) is tantamount to saying that any change in nominal interest rate must either cause or be directly linked to a corresponding and equal change in future equity cash flows. That simply seems hard for me to believe.

I think you acknowledge that fixed income is a viable investment alternative to equities. Yet, it seems that you assert (based on Asness' research?): that changes in real interest rates cannot change the NPV of future equity flows, because otherwise they must change the stock price and thus P/E; or, that real interest rates never change independent of future Equity Cash flows, so NPV will never change due to changes in real rates.
larryswedroe wrote:Isp

the CAPE would go up when the risk-free rate of return goes down.


Unfortunately, this is one of the most common mistakes people make, believing this should be true. Economists call it the money illusion. It's just not necessarily correct. From a piece I wrote on the subject


The belief that nominal interest rates should matter to stock valuations is based on the idea that bonds are competing instruments for stocks. But those that use metrics like comparing bond yields to P/E ratios to determine “fair value” are making an error by thinking of things in isolation; thus, they fail to see the whole picture. In this case, the belief is based on what is called the “money illusion.” It’s one that has great potential for creating mistakes because one of the most popular indicators used by investors to determine if the market is under- or overvalued, is The Fed Model.

In 1997, in his monetary policy report to Congress, Fed Chairman Greenspan indicated that changes in the ratio of prices in the S&P 500 to consensus estimates of earnings over the coming 12 months have often been inversely related to changes in long-term Treasury yields. Following this report, Edward Yardeni, speculated the Fed was using a model to determine if the market was fairly valued — how attractive stocks were priced relative to bonds. The model, despite no acknowledgement by the Fed, became known as the Fed Model.

Since Yardeni coined the phrase, the Fed Model as a valuation tool has become “conventional wisdom.” Unfortunately, much of conventional investment wisdom is wrong. There are two major problems with the Fed Model. The first is that the expected return of stocks isn’t determined by their relative value to bonds. Instead, the expected real return is determined by the current dividend yield plus the expected real growth in dividends. To get the estimated nominal return you add estimated inflation. This is a critical point that seems to be lost on many investors. The result is investors who believe low interest rates justify a high valuation for stocks without the high valuation impacting expected returns are likely to be disappointed. When P/Es are high, expected returns are low, and vice versa, regardless of the level of interest rates.

The second problem with the Fed Model is it fails to consider that inflation impacts corporate earnings differently than it does the return on fixed income instruments. Over the long term, the nominal growth rate of corporate earnings has been in line with the nominal growth rate of the economy. Similarly, the real growth rate of corporate earnings has been in line with the real growth of the economy. Thus, in the long term the real growth rate of earnings is not impacted by inflation. On the other hand, the yield to maturity on a 10-year bond is a nominal return — to get the real return you must subtract inflation. The error of comparing a number that isn’t impacted by inflation to one that is leads to what is called the “money illusion.” Let’s see why.

To keep our example simple we’ll assume that there’s no risk premium in the yield on nominal bonds for unexpected inflation and there’s no liquidity premium in the yield on TIPS (Treasury Inflation-Protected Security). Let’s also assume the real yield on a 10-year TIPS is 2 percent (wouldn’t that be nice), and the expected long-term rate of inflation is 3 percent. Thus, the 10-year Treasury bond would be 5 percent and the fair value for stocks would be at a P/E of 20. Now lower our assumption of inflation of 3 percent to 2 percent. This would cause the yield on the 10-year bond to fall from 5 to 4 percent, causing the fair value P/E to rise to 25. But, since inflation doesn’t impact the real rate of return demanded by equity investors, it shouldn’t impact valuations. In addition, as stated above, over the long term there is a strong relationship between nominal earnings growth and inflation. In this case, a long-term expected inflation rate of 2 percent, instead of 3 percent, would be expected to lower the growth of nominal earnings by 1 percent, but have no impact on real earnings growth (the only kind that matters).
Because the real return on bonds is impacted by inflation while real earnings growth is not, the Fed Model compares a number that is impacted by inflation with a number that isn’t — causing the money illusion. But, over time the P/E would in fact go up, because the E would be growing more slowly. So, if P stays the same (as you state), but E grows more slowly (lower inflation, so lower nominal earnings), the P/E must go up over time, no?

Now let’s consider what would happen if the real interest rate component of bond prices fell. The real rate is reflective of the economic demand for funds. Thus, it’s reflective of the rate of growth of the real economy. If the real rate falls due to a slower rate of economic growth, interest rates would fall, reflecting the reduced demand for funds. Using the earlier example, if the real rate on TIPS fell from 2 percent to 1 percent, that would have the same impact on nominal rates as a 1 percent fall in expected inflation, and, thus, the same impact on the fair value P/E ratio — causing fair value to rise. However, this too doesn’t make sense. A slower rate of real economic growth means a slower rate of real growth in corporate earnings. Thus, while the competition from lower interest rates is reduced, so will be future earnings. It seems to me that there is a assumption here that any change in real interest rates must be linked 1:1 with a change in future profitabilty/cash flows. I don't know that I would accept that assumption.

Since corporate earnings have grown in line with nominal GNP growth, a 1 percent lower long-term rate of growth in GNP would lead to 1 percent lower expected growth in corporate earnings. The “benefit” of falling interest rates would be offset by the equivalent fall in future expected earnings. The reverse would be true if a stronger economy caused a rise in real interest rates. The negative effect of a higher rate of interest would be offset by a faster expected growth in earnings. The bottom line is there is no reason to believe stock valuations should change if the real return demanded by investors hasn’t changed. But, are all changes in 'real' interest rates solely due to or directly linked to expectations of future expected earnings? What about changes in aggregate savings and consumption patterns, whether due to demographic changes or changing societal norms and mores?

When Clifford S. Asness studied the period 1881–2001, he concluded the Fed model had no predictive power in terms of absolute stock returns — the conventional wisdom was wrong. Asness also concluded that over 10-year horizons the E/P ratio does have strong forecasting powers. Thus, the lower the P/E ratio, the higher the expected returns to stocks, regardless of the level of interest rates, and vice versa. Interesting; but, how did he account for whether the changes in interest rates was due to changed 'real' interest rates or due to changes inflation expectations? There was no "Tips" marketplace for most of that time to gauge inflation expectations. Moreover, I am not saying that E/P has no value; what I am saying is that I just don't believe that changes in interest rates never affect the NPV of future equity cash flows, and thus that they must, at least sometimes, affect stock prices.

There’s one other point to consider. A stronger economy, leading to higher real interest rates, should be expected to lead to a rise in corporate earnings. The stronger economy reduces the risks of equity investing. In turn, that could lead investors to accept a lower risk premium. Thus, it’s possible that higher interest rates, if caused by a stronger economy and not higher inflation, could actually justify higher valuations for stocks. The Fed Model, however, would suggest that higher interest rates means stocks are less attractive. The reverse would be true if a weaker economy led to lower real interest rates. Again, it seems that the underlying assumption is that all changes in real interest rates are a function of or directly linked to the market's assessment of future economic growth. Let me turn that around: if real interest rates are so low right now (they are, aren't they?), doesn't that imply that future profit growth will be weak? In turn, wouldn't that increase the market's risk premium? So, weak profit growth and higher risk. If that is the case, why are the CAPE numbers so high?

larryswedroe
Posts: 15404
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Valuation does not matter (may be)

Postby larryswedroe » Fri Feb 17, 2017 9:35 am

ISP
Re lower real rate, and impact on NPV. As I explained, if real rate falls due to lower GNP growth and thus lower corporate earnings you get offsetting effects, and same thing going the other way. The lower rate in the denominator the NPV but the lower earnings in the numerator lowers the NPV

Second, can the real rate change for other reasons. Yes, as investors demand larger or small risk premiums for unexpected inflation for example. So flooding the market with money not only raises the nominal rate perhaps due to higher expected inflation the market might require a bigger risk premium for UNEXPECTED inflation. Just one example

Third, the Fed's view of course can differ from the markets but they are influenced by the market's view. And ultimately the FED only controls ST rates and not LT, and even there cannot fully control them. For example if try to lower rates by flooding market with greater money supply growth then you get inflation and rates rise no matter what the FED does.

Fourth, I don't agree at all with your conclusion because, as I explained, all things are NOT equal. Now investors are unfortunately it seems fooled by this money illusion, and that impacts stock prices which are hard to correct because of limits to arbitrage and costs/risks of shorting. So you can get bubbles.

Fifth I don't agree at all that changes in real rates cannot impact stocks, that IMO is clearly wrong. Example, FED tightens monetary policy, driving real ST rates way up and kills economic growth and stock prices crash for two reasons. The numerator of the NPV falls as corporate earnings forecasts fall and the denominator goes up as the riskless rate goes up.

But, over time the P/E would in fact go up, because the E would be growing more slowly. So, if P stays the same (as you state), but E grows more slowly (lower inflation, so lower nominal earnings), the P/E must go up over time, no?

No, and don't know why you think that is so.

It seems to me that there is a assumption here that any change in real interest rates must be linked 1:1 with a change in future profitabilty/cash flows. I don't know that I would accept that assumption.
That doesn't of course have to be case, just showing how it can happen as in the example. Of course the RISK PREMIUM in the real rate, meaning the risk of unexpected inflation, can move as well. Also as noted rising real rates if caused by FED and not demand can raise the risk premium demanded by investors, and vice versa. And hard to disentangle the effects. And that answers your next question

Re Asness and estimate of inflation, don't have time to dig back in that paper but what people do is estimate the market's inflation estimate by using lagged inflation.

Again, it seems that the underlying assumption is that all changes in real interest rates are a function of or directly linked to the market's assessment of future economic growth. Let me turn that around: if real interest rates are so low right now (they are, aren't they?), doesn't that imply that future profit growth will be weak? In turn, wouldn't that increase the market's risk premium? So, weak profit growth and higher risk. If that is the case, why are the CAPE numbers so high?

Again, as I explained there can be other effects, the examples meant to show why assuming lower rates mean higher P/E, which isn't true necessarily. That's the money illusion problem.

The CAPE is high for many reasons, some of which I addressed in the paper I wrote, including changes in FAS 142 and 144, lower dividend payouts which assume then higher future growth of earnings, lower implementation costs, country is wealthier so capital less scarce (so costs less---note all around the world there is negative relationship between wealth and erp). And also the FED has accomplished it's objective of pushing people to take more risks because they cannot meet their goals with 0 fixed income returns and you get people making mistakes like thinking dividend paying stocks are substitutes for safe bonds. And of course the market perceives the risks of economic downturn to be small. And one could add other reasons.

Hope helpful,
Larry

Alchemist
Posts: 156
Joined: Sat Aug 30, 2014 6:35 am
Location: Florida

Re: Valuation does not matter (may be)

Postby Alchemist » Fri Feb 17, 2017 11:29 am

A bit of the detail on interest rates vs P/E effects is admittedly a bit above my head. So for others like me, I will throw out an important reminder. Valuations (and other aggregate metrics) are the forest. Real companies making/selling real things are the trees. Valuations can indeed go down because prices fall (market correction). But also remember than can go down because earnings increase, leading to long running bull markets.

No one knows which will happen in the future. So stay the course and try not to get too worked up over the noise. That is not mean to discourage any of the discussion going on here in this thread. It's all pretty interesting stuff and I (and certainly others) are learning a great deal from it. Rather, just trying to make sure the important things are kept in perspective.

lsp12
Posts: 18
Joined: Thu Dec 29, 2016 12:20 pm

Re: Valuation does not matter (may be)

Postby lsp12 » Fri Feb 17, 2017 6:48 pm

larryswedroe wrote:

The CAPE is high for many reasons, some of which I addressed in the paper I wrote, including changes in FAS 142 and 144, lower dividend payouts which assume then higher future growth of earnings, lower implementation costs, country is wealthier so capital less scarce (so costs less---note all around the world there is negative relationship between wealth and erp). And also the FED has accomplished it's objective of pushing people to take more risks because they cannot meet their goals with 0 fixed income returns and you get people making mistakes like thinking dividend paying stocks are substitutes for safe bonds. And of course the market perceives the risks of economic downturn to be small. And one could add other reasons.

Hope helpful,
Larry


Just a couple more thoughts: it seems to me part of the quoted paragraph reflects the underlying premise of the CAPM. If fixed income (LTGB) goes to 0, then the risk free rate of return is 0. Thus, people value the future returns of stocks more highly, because when real rates fall, NPV goes up, driving up the price of stocks, thus a higher P/E. Now, you can say people are being naive, stupid, mis-led, underestimate the risks, whatever, but the bottom line is, lower real interest rates, higher stock prices, higher P/E. You seem to assert that the market has reduced the required risk premium instead, and that is what is driving higher P/E. I don't know, perhaps. Just doesn't seem likely to me; why would lower rates truly increase people's risk tolerance or affect the market's assessment of equity risk? Do you really believe that, in aggregate, the market is that stupid/naive/ill-informed? Isn't it more likely to have bid up P because the NPV of future earnings has increased? Either the market is smart, or it isn't. I don't see how changes in real interest rate levels would cause the market's aggregate level of 'intelligence' to change.

Moreover, earlier, you used the example of falling rates implying lower future GDP growth, and that in turn INCREASING the risk associated with equities. So, which is it: lower rates, lower risk premiums, and thus higher equity prices; or lower rates, higher risk premiums, and thus lower equity prices?

If someone says: lower interest rates and higher equity prices (higher P/E), ergo people's risk premium has fallen; that is making a HUGE, and unstated assumption.

Thus, the other thing that I conclude from your comments is that it seems -- to me -- that you are essentially asserting something, then using that assertion to prove your point. It feels to me like circular logic or arguing a tautology based on a limiting assumption. What I mean is: if rates rise (for example), it is either: a. due to higher inflation expectations; b. due to expectations linked to expectations for higher GDP growth and thus greater demand for capital/a view that GDP prospects have improved and that will flow through to profits, or c. some other reason. You seem to dismiss (c) as a possibility.

So, I will grant you that, if only a or b could be the reason that rates rise, then there is no reason that equity prices should change. However, all of your arguments seem to only allow a or b, and of course if you assume or assert that only a or b are possible, then by definition P/E doesn't change when interest rates change, since, in real terms, NPV of E isn't changing, thus nor would P. But, you implicitly reject possibility c. If c does happen (that real interest rates are (sometimes) decoupled from future prospects/expectations of GDP growth and associated real earnings growth), then changes in interest rates which differ from GDP/profit growth expectations must drive changes in NPV of future earnings, which will drive changes in stock prices, which will cause changes in P/E. So, here is a question: how do real interest rates correlate with expectations of future real GDP growth? It seems that you assert that they are linked 1:1; otherwise, as least sometimes, changes in real interest rates must change P/E ratios, unless somehow that change is affecting the required risk premium. And, parenthetically, it isn't fair to measure expectations of future GDP growth based on stock prices, because that is assuming that the two are in fact completely linked.

I appreciate the dialogue.

User avatar
willthrill81
Posts: 742
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Valuation does not matter (may be)

Postby willthrill81 » Fri Feb 17, 2017 7:06 pm

The chart below shows the 25 year rolling averages of the CAPE-10 (using GAAP data, which isn't as good as the NIPA data, but still useful) from 1905 to 2017. Yes, I know that CAPE itself is an average of the current S&P 500 price divided by the previous 10 years earnings, so this is an 'average of the average', but I still think it's interesting, if nothing else.

Image

This shows that the 25 year average was comparatively high from 1905-1912, then trailed downward until around 1929. Then, apart from a downturn from 1951-1960, it moved steadily upward until 1974, when it began a slow decline until 1993. Since then, it has been moving steadily upward.

During the 113 years of data in this graph, there have been a total of three periods where the 25 year average steadily declined (1912-1929, 1951-1960, 1974-1993). Outside of these periods, the 25 CAPE average has been on an upward trend.

I think this is important. Just looking at each year's CAPE, this trend is not so visible. But looking at 25 year averages, it becomes clear. This made me interested, so I looked at the correlation between CAPE and time, finding it to be .55, a moderately strong, positive relationship. This means that time alone is accounting for around 30% of the variation in CAPE, and CAPE is going up over time. While the 25 year CAPE average is at its all time high, this is not so alarming when you see that it's been hitting a new high every year since 2000, although its growth rate has slowed since 2007.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

larryswedroe
Posts: 15404
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Valuation does not matter (may be)

Postby larryswedroe » Fri Feb 17, 2017 7:32 pm

ISP
First, you keep repeating, incorrectly, the assertion that I think it's the only explanation. I repeat, falling real rates can reflect several different things, including lower economic growth, which should mean lower corporate earnings and thus offsetting the impact of lower rates or it could be lower risk premium required by investors. Most, including those using the so called Fed Model, which you are using, ignore completely the other explanation, which is more likely, more often.

Second,
If fixed income (LTGB) goes to 0, then the risk free rate of return is 0. Thus, people value the future returns of stocks more highly, because when real rates fall, NPV goes up, driving up the price of stocks, thus a higher P/E.
Here again you repeat the same error. Real rates going to zero could reflect forecasts of very poor economic growth and poor earnings outlook and that should not mean stock prices rise. Note real rates have been historically in line with both real gnp growth and real corporate earnings growth. So if real rates fall...... the assumptions should follow. Now this isn't always the case, particularly after a Fed tightening to fight inflation and then rates start to fall, as they did in the 1980s.

Note I am only asserting not only standard economic theory but the factual evidence, as I just showed above about the relationship between real rates and economic growth.

I urge you to read Cliff Asness's piece here from the journal of portfolio managementhttp://faculty.som.yale.edu/jakethomas/otherpapers/Asness.pdf

Larry

lsp12
Posts: 18
Joined: Thu Dec 29, 2016 12:20 pm

Re: Valuation does not matter (may be)

Postby lsp12 » Fri Feb 17, 2017 9:42 pm

larryswedroe wrote:ISP
First, you keep repeating, incorrectly, the assertion that I think it's the only explanation. I repeat, falling real rates can reflect several different things, including lower economic growth, which should mean lower corporate earnings and thus offsetting the impact of lower rates or it could be lower risk premium required by investors. Most, including those using the so called Fed Model, which you are using, ignore completely the other explanation, which is more likely, more often.

Second,
If fixed income (LTGB) goes to 0, then the risk free rate of return is 0. Thus, people value the future returns of stocks more highly, because when real rates fall, NPV goes up, driving up the price of stocks, thus a higher P/E.
Here again you repeat the same error. Real rates going to zero could reflect forecasts of very poor economic growth and poor earnings outlook and that should not mean stock prices rise. Note real rates have been historically in line with both real gnp growth and real corporate earnings growth. So if real rates fall...... the assumptions should follow. Now this isn't always the case, particularly after a Fed tightening to fight inflation and then rates start to fall, as they did in the 1980s.

Note I am only asserting not only standard economic theory but the factual evidence, as I just showed above about the relationship between real rates and economic growth.

I urge you to read Cliff Asness's piece here from the journal of portfolio managementhttp://faculty.som.yale.edu/jakethomas/otherpapers/Asness.pdf

Larry


I certainly don't claim absolute certainty about any of this. If you think I am, I apologize.

I find the conversation interesting, and I will read the Asness piece.

That being said, in your last reply you use terms such as 'can reflect,' 'including,' 'should,' and 'this isn't always the case.' Those are inherently words that retain the option of saying "I never said it could never be the case, or that I am saying I absolutely disagree with your assertion." They leave open the option to later say 'in some instances, you were right, I never said you couldn't be right.' I have a friend who would call them weasel words (and please know, I mean no offense when I say that; this is a purely intellectual conversation, and I take none of it nor mean any of my words as a personal attack). As a graduate of law school, I would say they are lawyerly words (though my wife would say they are pedantic).

However, my point being this. Based on your reply to my first post on the possible relevance of interest rates to CAPE, it seems that you reject, outright, any impact of interest rates on CAPE. Now, your last reply seems to hedge your position, using qualifiers and hedging your assertions.

In other words, are you, as you seem to in your reply to my first post, rejecting absolutely the possible impact of changing interest rates on P/E, or are you saying that you don't find it to be always a cause, but do recognize the possibility or episodic impact of changing interest rates on P/E? Based on my intellectual framework and understanding, I don't agree with the former and would tend to accept the latter (and thus, by extension, acknowledge that changing interest rates alone won't explain all changes in P/E).

Or, to my original point, I find the analysis of CAPE, independent of interest rate, to be intellectually incomplete. I am not saying I know exactly how to improve it, given that interest rates include a 'real' and an inflation component, just that I don't believe that interest rates (at least real interest rates) can be ignored when looking at CAPE. However, I never see the issue of underlying (real) interest rates raised when CAPE is analyzed, which I think ought to be at least addressed.

lsp12
Posts: 18
Joined: Thu Dec 29, 2016 12:20 pm

Re: Valuation does not matter (may be)

Postby lsp12 » Sat Feb 18, 2017 8:34 am

I have read the Asness article, though I have not delved in to all the details.

However, let me cite three quotes from the article regarding interest rates and P/E's:

"While the Fed model fails as a predictive tool for future long-term stock returns, it does work as a descriptive tool for how investors choose to set current stock market P/Es."

"Note that this finding that the Fed model has descriptive power for how investors set P/Es in no way contradicts the finding that the Fed model fails as a predictive tool for stock returns."

"There is strong evidence that investors set stock market E/Ps lower (P/E’s higher) when nominal interest rates are lower (and vice versa)."



Here is the first sentence from my initial post in this thread:
"One of my issues with Schiller and CAPE is that it doesn't account for changes in interest rates/discount rate (as far as I know). "

In other words, my point is this: when looking at CAPE, if the current CAPE is (making up numbers in all instances) 22, and the historical average is 15, it seems fallacious to me to say CAPE is too high, or argue for a reversion to the mean, if the reason current CAPE is at 22 is because the risk free rate has fallen to 0 whereas the average CAPE of 15 was in environments where the risk free rate was, say, 2%. That is all I am saying -- I never see a discussion of current risk-free rates, or comparisons thereof, in a discussion of the Schiller numbers.

I believe we've been talking past each other. I never intended to suggest that interest rates should be used to predict future returns of stocks. All I said was that I don't understand how, when people assess CAPE/Schiller, and they look at the current CAPE as compared to historical CAPE, they never seem to relate those comparisons to where interest rates are and have been. I believe the first quotation validates my point: if the market is setting P/Es in part based on current interest rates, then how can one assess today's CAPE and how it compares to historical CAPE without looking at interest rates?

You took my question/observation and turned it into a discussion of the errors of using the Fed model to forecast future equity returns. I never mentioned the Fed model, nor did I suggest that comparing P/E to Y was a useful equity forecasting tool. I make no suggestions whatsoever about the power of various forecasting or analytical tools for assessing future stock returns. All I am saying is that I don't understand why prevailing interest rates are never part of the Schiller/CAPE discussion.

staythecourse
Posts: 4312
Joined: Mon Jan 03, 2011 9:40 am

Re: Valuation does not matter (may be)

Postby staythecourse » Sat Feb 18, 2017 9:23 am

I am not sure why we always debate academic points as if any of us are writing a PHD paper.

The question the individual investor should be asking is NOT does valuations matter, but is there a set of metrics that give better returns on a prospective basis after transaction cost, taxes, brokerage costs, etc... then the default option of "oh well I'll just throw money into my static asset allocation each month" over even a 25 year time horizon (half of an investor's life). If the answer is not a resounding yes then who cares about the academic arguments as the only logical approach is to choose the default option.

In my opinion, does valuations matter? Sure I'm sure it does. Does it help investors in the real world improve returns or control risk? No I think the data does not support it. The only time I have seem valuations make a difference is when the market goes up or down and folks are trying to explain why and sound smart doing it vs. "Heck I have no clue".

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle

larryswedroe
Posts: 15404
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Valuation does not matter (may be)

Postby larryswedroe » Sat Feb 18, 2017 9:44 am

ISP
Sorry but you are completely misinterpreting what I have said and it has NOTHING to do with weasel words. It's just that you keep failling to understand that it's not black or white, but that there are many things going on and it's hard to parse what is driving things. So my LAST simple example for you and hope it helps clarify

Real interest rates fall because economic demand falls, and with slower economy corporate earnings fall. The fall in real rates can help stocks as the discount rate in the denominator falls but what you miss is that the earnings in the numerator also falls. That lowers prices. And the weaker demand can also have impact on the risk premium, added to the risk free rate, that investors demand, and that is part of the denominator as well. No way to know what will happen. And a good example of why that is the case is that the FED's ZRP has driven many cash flow investors to shift their risk preferences, buying say dividend paying stocks as substitutes for safe bonds. So that impacts the denominator in the reverse way, offsetting some or all of the impact from weaker economy.

And finally, you are wrong in saying that interest rates are never part of the discussion about the CAPE 10, it's in fact often discussed because of the issues I raised, how it can shift investor behaviors, and because of the money illusion issue.

Larry

minimalistmarc
Posts: 85
Joined: Fri Jul 24, 2015 4:38 pm

Re: Valuation does not matter (may be)

Postby minimalistmarc » Sat Feb 18, 2017 9:58 am

Is it probable that any analysis of stock market history could prove futile as nobody can predict the future and history may not repeat.

I prefer the grammatically incorrect "Nobody knows nuthin'!"

larryswedroe
Posts: 15404
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Valuation does not matter (may be)

Postby larryswedroe » Sat Feb 18, 2017 10:18 am

minimalist
It's not that we don't know anything, we do know a lot, but we don't know the future, what events will occur to cause the risk premium to change, how the economy will perform which determines earnings, etc.
Larry

lsp12
Posts: 18
Joined: Thu Dec 29, 2016 12:20 pm

Re: Valuation does not matter (may be)

Postby lsp12 » Sat Feb 18, 2017 11:23 am

larryswedroe wrote:ISP
Sorry but you are completely misinterpreting what I have said and it has NOTHING to do with weasel words. It's just that you keep failling to understand that it's not black or white, but that there are many things going on and it's hard to parse what is driving things. So my LAST simple example for you and hope it helps clarify Perhaps I am misinterpreting what you've written. I certainly feel you've misinterpreted what I've written; maybe you will go back and re-read my comments.

I never said anything was black and white, nor would I suggest that it is. Of course, anyone with any even marginally sophisticated understanding of the market knows that stock prices reflect all sorts of things: risk free rates of return, perceived risks, required risk premiums, expectations of earning growth, etc. etc. etc. By extension, a change in any of these will affect stock prices. Where did I ever say it is simple? If it were simple, why would there be all the ongoing debate and academic research into the topic of market valuation, etc.? All I am saying is that surely the risk free rate of return needs to be one of the things in the conversation when Shiller ratios are discussed, since surely it can be a factor in P/Es.


Real interest rates fall because economic demand falls, and with slower economy corporate earnings fall. The fall in real rates can help stocks as the discount rate in the denominator falls but what you miss is that the earnings in the numerator also falls. That lowers prices. And the weaker demand can also have impact on the risk premium, added to the risk free rate, that investors demand, and that is part of the denominator as well. No way to know what will happen. And a good example of why that is the case is that the FED's ZRP has driven many cash flow investors to shift their risk preferences, buying say dividend paying stocks as substitutes for safe bonds. So that impacts the denominator in the reverse way, offsetting some or all of the impact from weaker economy. Let me reiterate: You now admit (see bolded portion of your comment above) that a higher P/E could be the result of a higher NPV based on a lower risk free rate rather than a lower risk premium. Plus, Asness, whom you introduced into the discussion to prove your point, states that interest rates do affect P/Es! Again, that is all I'm saying (see previous paragraph "All I am saying is that surely the risk free rate of return needs to be one of the things in the conversation when Shiller ratios are discussed, since surely it can be a factor in P/Es.", and, not to beat a dead horse, but my very first sentence in this thread: "One of my issues with Schiller and CAPE is that it doesn't account for changes in interest rates/discount rate (as far as I know). "). Nothing more!

And finally, you are wrong in saying that interest rates are never part of the discussion about the CAPE 10, it's in fact often discussed because of the issues I raised, how it can shift investor behaviors, and because of the money illusion issue. Well, in my -- admittedly less than exhaustive -- reading of discussions of Schiller etc, I don't recall seeing it. That was my original point, before we got into such a long-winded dialogue. So, if you can point me to some nuanced discussion of Shiller and CAPE that incorporates a discussion of the role of the Risk Free Rate, I would be interested in reading that discussion. Does Shiller ever incorporate that into his analyisis? Thanks in advance for that!

Larry


Let me play back our discussion, as I see it:

LSP: discussions of Shiller/CAPE ought to include the topic of changes in risk free rates of return when looking at P/E ratios across/between time periods
Larry: The Fed model, which LSP is referring to, is flawed, and changes in interest rates do not affect P/E (citing Asness)
[
quoting LSP: "the CAPE would go up when the risk-free rate of return goes down. "



Larry: "Unfortunately, this is one of the most common mistakes people make, believing this should be true. Economists call it the money illusion. It's just not necessarily correct."]

LSP: I don't understand how changes in the risk free rate can't/don't affect P/E's
Larry: let me explain the money illusion and investors' flawed logic
LSP: wait a minute, having actually read the article, Asness in fact does acknowledge that interest rates do affect P/Es
Larry: well yes, but the market is complicated, so LSP is being overly simplistic
LSP: All I'm saying is risk free rates can affect P/Es, so discussions of Shiller and CAPE should incorporate the topic of real interest rates and their effect on the ratios.
Last edited by lsp12 on Sat Feb 18, 2017 11:37 am, edited 1 time in total.

User avatar
packer16
Posts: 738
Joined: Sat Jan 04, 2014 2:28 pm

Re: Valuation does not matter (may be)

Postby packer16 » Sat Feb 18, 2017 11:25 am

Although the argument is true about lower interest rates = lower expected growth this assumes the economy everywhere has is run with the same rules & there are no significant flows from another economy due to property rights differences. Today, there are different property right and expropriation risks in China and other emerging markets. The flow or funds into the developed economies is not based only upon expected growth but expected protection of property rights. In most economic models, the assumption is property rights are the same everywhere but in reality this is not the case. I think one of the reasons for low interest rates in developed markets is this flow in addition to low economic growth. I think this flow will continue & will offset in part higher rates due to higher economic growth.

Packer
Buy cheap and something good might happen

mikeguima
Posts: 66
Joined: Mon Aug 22, 2016 3:11 pm

Re: Valuation does not matter (may be)

Postby mikeguima » Sat Feb 18, 2017 12:27 pm

larryswedroe wrote:will, fwiw, IMO that 1% figure is crazy wrong. Not even close to reality. And the people I talk to at high levels in the field would agree.
The ERP has historically ben 1%. I don't know anyone who thinks it is now 7% Stocks historically have returned about 7% real and most I talk to think somewhere between 4-5%, and that assume NO reversion to mean of valuations. If we do get that obviously returns will be much lower.
Wonder if the person giving this advice was saying in 2000 that should adjust returns no more than 1%? (Rhetorical)
Larry

Only 1% historical equity risk premium? I've always read it was higher, averaging around 4% (according to Siegel in his book, Stocks for the long run). Hence the equity risk premium puzzle, meaning the premium for investing in stocks is higher than can be explain by added risk relative to treasuries. Just 1% doesn't seem like it would justify the puzzle as it probably would be a fair ERP.

User avatar
willthrill81
Posts: 742
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Valuation does not matter (may be)

Postby willthrill81 » Sat Feb 18, 2017 12:39 pm

mikeguima wrote:
larryswedroe wrote:will, fwiw, IMO that 1% figure is crazy wrong. Not even close to reality. And the people I talk to at high levels in the field would agree.
The ERP has historically ben 1%. I don't know anyone who thinks it is now 7% Stocks historically have returned about 7% real and most I talk to think somewhere between 4-5%, and that assume NO reversion to mean of valuations. If we do get that obviously returns will be much lower.
Wonder if the person giving this advice was saying in 2000 that should adjust returns no more than 1%? (Rhetorical)
Larry

Only 1% historical equity risk premium? I've always read it was higher, averaging around 4% (according to Siegel in his book, Stocks for the long run). Hence the equity risk premium puzzle, meaning the premium for investing in stocks is higher than can be explain by added risk relative to treasuries. Just 1% doesn't seem like it would justify the puzzle as it probably would be a fair ERP.


To be fair, Larry was talking about a different time period than Kitces, who demonstrates quite simply that even when valuations are high, the impact on the the 30 year average returns is small, around a 1% reduction, so 5.5% real annual growth. It has higher explanatory power for the following 15 years of returns, and Kitces shows that those returns can be 4% lower than the average (~6.5% real average, so potentially 2% following very high valuations), but this usually leads to greater than average growth in the following 15 years.

Keep in mind though that CAPE can explain no more than 40% of the variation in stocks. That means that the majority is due to other factors, and anything can happen.
Last edited by willthrill81 on Sat Feb 18, 2017 4:07 pm, edited 1 time in total.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

larryswedroe
Posts: 15404
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Valuation does not matter (may be)

Postby larryswedroe » Sat Feb 18, 2017 1:42 pm

ISP
I have ALWAYS said a higher CAPE can be result of lower ERP, don't know how you thought otherwise.
That just means expected returns are LOWER, not overvalued. I even wrote an article on that subject discussing the idea that the CAPE 10 was saying market overvalued, ala Grantham and Hussman. I even provided logical explanations for why the risk premium should have fallen which can have nothing to do with interest rates. But also when real rates fall investors can be "pushed" to take more risk and valuations rise, and thus future returns go down.

The fact is we don't know exactly why real rates move, it could be lower economic demand, could be Fed actions, and can lead to investors taking more risk, and some combinations. But it still means lower expected returns when CAPE 10 rises, but doesn't mean market overvalued.

Thus IMO you can basically ignore rates, and if they mattered all that much btw, someone would have built a model that incorporated it and improved the forecasting power of the CAPE--with all that computing power and the money to be made, you'd think someone would have done it. Including Shiller. But perhaps someone will. Note Asness doesn't adjust CAPE 10 in his forecast of returns. And I don't know of anyone else that does, though doesn't mean some don't. So no I cannot point you to any discussion on it.

larryswedroe
Posts: 15404
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Valuation does not matter (may be)

Postby larryswedroe » Sat Feb 18, 2017 1:44 pm

ISP
I have ALWAYS said a higher CAPE can be result of lower ERP, don't know how you thought otherwise.
That just means expected returns are LOWER, not overvalued. I even wrote an article on that subject discussing the idea that the CAPE 10 was saying market overvalued, ala Grantham and Hussman. I even provided logical explanations for why the risk premium should have fallen which can have nothing to do with interest rates. But also when real rates fall investors can be "pushed" to take more risk and valuations rise, and thus future returns go down.

The fact is we don't know exactly why real rates move, it could be lower economic demand, could be Fed actions, and can lead to investors taking more risk, and some combinations. But it still means lower expected returns when CAPE 10 rises, but doesn't mean market overvalued.

Thus IMO you can basically ignore rates, and if they mattered all that much btw, someone would have built a model that incorporated it and improved the forecasting power of the CAPE--with all that computing power and the money to be made, you'd think someone would have done it. Including Shiller. But perhaps someone will. Note Asness doesn't adjust CAPE 10 in his forecast of returns. And I don't know of anyone else that does, though doesn't mean some don't. So no I cannot point you to any discussion on it.


Return to “Investing - Theory, News & General”

Who is online

Users browsing this forum: alec, AlohaJoe, Bing [Bot], freyj6, Hodor, ishallnotwant, nybble, packet, segfault, snowshoes, Wakefield1, white_water and 85 guests