Can someone convince me we aren't in for a decade or two of no return?

[vinod1]

Philosophical economics has talked about why the costs, etc in historical context beautifully in one of the blog posts.

 

http://www.philosophicaleconomics.com/

 

Vinod

[vinod1]

A lot of popular investors implicitly assume this 6.5% real return as the divine right of investors. This includes Montier, Hussman among others.

 

Vinod

[SHDL]

But the fact remains that we have less than 150 years of stock market data. If you believe long term to be 20 to 30 years, then we have non-overlapping sample size of 5 or 7.

 

Hardly the amount of data that you could use to draw conclusions. Maybe in another 1000 to 2000 years we would be a better position.

 

But to me, at this time with the data we have, to draw strong conclusions about market valuation based on such short data set seems to be unwise.

 

This is very true.  In fact one could even argue that since we only have one world economy, the sample size is (and will forever be) 1!

 

The way I see the discussion here (and elsewhere) is that we're mostly just a bunch of (quasi-)Bayesian learners talking about their subjective beliefs.  The data limitations you mention is one of the fundamental reasons why our beliefs don't always agree.

[Castanza]

You cannot talk about expected returns without factoring in required returns. It has a big impact on valuations.

 

Most just assume that since the realized returns were around 6.5% real over the last century or so, that is also what investors would/should require going forward.

 

Think back to the time around 1900, people started hearing about Dow Jones averages for 5 to 15 years at most. They had very little data on the stock market and not much information to form an opinion on what returns to expect.

 

Most importantly to put together a portfolio of 20 to 30 stocks would have been very expensive that consumed a significant portion of the expected returns.

 

So the 6.5% real returns that investors realized would have been more like 4% or 5% real, because of expenses (transaction costs, fraud, cost to physically secure the paper certificates, etc).

 

Now we have practically cut down on the expenses to zero and a widely diversified portfolio can be put together easily at the click of a button. I would think that probably reduced the actual cost paid by the investors by 2% or so compared to the 1900s.

 

In addition, I would also argue that the economic/political risks are much less than in the past. So that would require even less of required return.

 

If you put all of these together, I think the required real return for investors is going to be much less than in the past. I do not know how much, but 4% real does not seem to be too outlandish to me.

 

Put this into the valuation and you might get a much less scary picture of the valuation.

 

But the fact remains that we have less than 150 years of stock market data. If you believe long term to be 20 to 30 years, then we have non-overlapping sample size of 5 or 7.

 

Hardly the amount of data that you could use to draw conclusions. Maybe in another 1000 to 2000 years we would be a better position.

 

But to me, at this time with the data we have, to draw strong conclusions about market valuation based on such short data set seems to be unwise.

 

Vinod

 

Even if we were to wait 1000 years for better sample sizes how accurate would the data even be? I'm not a fan of historical averages. I mean look at how the world has changed in the last 50 years. There are far too many variables in my opinion to draw any type of conclusion which would make me feel confident in averaging a 7% return. The time is now and whatever the return now is what we will get. I think putting more effort into real time vs historical time is far more important. But there still is value in terms of historical perspectives. Especially when it comes to policies, and psychological/sociological trends (Which can be hard to justify at times as well.)

 

I'll leave this skyline montage to simply show how much places in Asia, SA, and the Middle East have changed. Certainly these have effects on today's economy compared to the early 1900's when the populations were low all over and countries were more nationally focused.

 

http://the-technology-tips.blogspot.com/2014/03/20-skylines-of-world-then-vs-now.html

[Cigarbutt]

It's fair to say that looking to buy or sell the market likely isn't a useful exercise most of the times, especially looking bottom-up.

But a lot of the assumptions seem to be taken for granted and rare events occur much more frequently than statistical models would suggest.

 

Mr. Buffett once quipped that he would be ready to pay a large sum in order to know if all his stock holdings would fall by 50% within a month but he has also said that investors in the stock market should be ready to handle an across the board 50% decline.

 

So, there is a price to pay for protection and a potential price to pay for unexpected reactions.

 

Even if historical comparisons are only a starting point, I find that the late 60's share a lot of features with now (but there are also huge differences) in terms of overall valuations and outlook. Interesting because then, it looked like Mr. Buffett preferred bonds over equities for a while which is markedly different from now although he has kept the financial flexibility constant.

 

[SharperDingaan]

Few things to add to this.

 

The more historic the data the less valuable it is for predictive purposes. Simply because todays economy is very different to what it was even 20 years ago, and we are not comparing apples to apples.

 

Agreed central bank currency inflation is highly likely to continue; but asset inflation and 'growth' is a real crap shoot.

Mathematically, the bigger you are the harder it is to get incremental growth; and most would suggest that for the forseeable future - the bulk of global growth will be in Asia, and not in NA or Europe. Most would also expect a run-off of NA/European assets in favour of a build-up of Asian ones. So what? The forecast for NA/Europe is essentially currency inflation +/- (asset inflation+growth); 2-3%+ nominal growth on a good day, or 0-?% after inflation. Charming.

 

Sniff test.

Aging NA boomers are now taking money retirement money OUT of their savings/retirement accounts; their net investments, houses, toys, etc. are now selling down, NOT being built up. So .... if the forecast bias is for ASIAN assets, and NA assets are persistently in a GROWING net sale position, NA asset prices must fall, and deflate. To avoid that deflation, NA central banks must print currency faster, & target higher inflation. Maybe 3% nominal growth, 0% after inflation, and never ending QE?

 

Point?

 

We may well see CYCLICAL equity returns of 12% plus;

but we're compounding year-over-year at maybe 2-3%/yr nominal, and 0-1% real.

 

SD

 

 

 

 

 

 

[SHDL]

Also, I have mentioned this before, I think the next stock market rout may well be caused by political events.

 

I'd expect a pretty sharp sell off if something like this were to go through:

 

https://www.wsj.com/articles/top-democrat-proposes-annual-tax-on-unrealized-capital-gains-11554217383?mod=hp_lista_pos3

[DTEJD1997]

Also, I have mentioned this before, I think the next stock market rout may well be caused by political events.

 

I'd expect a pretty sharp sell off if something like this were to go through:

 

https://www.wsj.com/articles/top-democrat-proposes-annual-tax-on-unrealized-capital-gains-11554217383?mod=hp_lista_pos3

 

I was SERIOUSLY considering RADICALLY modifying my investment strategy going forward.  I was considering switching over to income stocks, refraining from investing in long term projects, especially real estate, and generally raising cash.

 

In order for investing to work, you've got to have trust between generations.  You also have to have roughly the same set of rules.

 

I was not so sure that things would look roughly the same 5, 10, 15 years from now.

 

NOW, I am much more optimistic.

 

I think politically, things will start to calm down.

[CorpRaider]

Also, I have mentioned this before, I think the next stock market rout may well be caused by political events.

 

I'd expect a pretty sharp sell off if something like this were to go through:

 

https://www.wsj.com/articles/top-democrat-proposes-annual-tax-on-unrealized-capital-gains-11554217383?mod=hp_lista_pos3

 

Seems like the politicians should (likely will, imop) consider repealing the special rates for capital gains first and reinstate the former "wealth tax", which as Buffett recently noted was just referred to as the estate and gift tax.  Seems it would be so much simpler and easier to sell that than the new "wealth tax" and all these sur taxes on buybacks and junk.  Just roll back the capital gains preference and reinstate the estate tax....use something that has been used just fine before...not even that long ago....derp.

[cameronfen]

Ok I'm clearly in the minority, but let me add, from 1926 to 1946 which I think is at least among the worst performing 20 year period ( I'm on my phone  with no ability to use excel thoroughly) and the total return averaged 7%.  15 years return averaged 4%. People saying the returns for the next 15 to 20 years will be 5-7% nominal are basically saying the markets will perform as bad or worse than they did for the 20 year period covering the great depression. 

[SHDL]

Ok I'm clearly in the minority, but let me add, from 1926 to 1946 which I think is at least among the worst performing 20 year period ( I'm on my phone  with no ability to use excel thoroughly) and the total return averaged 7%.  15 years return averaged 4%. People saying the returns for the next 15 to 20 years will be 5-7% nominal are basically saying the markets will perform as bad or worse than they did for the 20 year period covering the great depression.

 

The S&P struggled a lot after the dot com bubble too.  Major valuation metrics like the CAPE ratio are similarly elevated, and so I don’t think the low expectations are crazy in light of the evidence.

 

One big difference though is that this time the case for selling stocks and moving into something like cash or Treasury bonds is a lot weaker because their yields are very low too.  Unless that changes it’s possible that we end up just sloooooowly grinding higher. 

[scorpioncapital]

But is it possible to do both at once? If rates are high enough that you want to buy treasuries and sell stocks, wouldn't your stocks be somewhat depressed by that point? catch 22?

[SHDL]

But is it possible to do both at once? If rates are high enough that you want to buy treasuries and sell stocks, wouldn't your stocks be somewhat depressed by that point? catch 22?

 

Sorry, I probably didn’t follow. 

 

What I meant by “slowly grinding higher” is an outcome where people end up with something like 2.5% nominal returns on cash/Treasuries and 4% nominal returns on the S&P over the long haul without experiencing major ups or downs.

 

If interest rates were to go up in a big way I think many people will try to move out of stocks and into bonds and that will make stock prices go down.

[Gregmal]

I also think the concern listed in the thread title is mainly relevant for those that just want to be lazy and index everything. If you own high quality, cash flow generating companies that buyback their own stock, supply and demand by itself will create your returns. If not, you are afforded the opportunity to buy shares at a depressed valuation. win/win. This obviously assumes nothing macro happens that effects said company's ability to generate fcf, or on a fundamental basis that causes the company's prospects to deteriorate. But as always those that do the work should continue to get rewarded.

[TwoCitiesCapital]

If you top ticked the S&P 500 in 2007 and held all the way down then back up to present, you'd have averaged something like 7.5% a year. 

 

I don't think stocks are nearly as expensive now as they were in 2007, so i'll throw my guess in the there and predict that stocks will do about 7 or 8% a year over the next decade. 

 

Also don't see a recession on the horizon at the moment, of course that involves some guess work on the intentions/future actions of the Fed, so take it with a grain of salt.

 

As cameronfed alluded to, multiples should be higher going forward. The standard average 16 P/E Ratio occurred when nominal G.D.P. averaged over 6% in the 20th century.

 

I think N.G.D.P. will be more like 4% going forward, so multiples will likely be higher...and current multiples are justified on a long term basis

 

Isn't this the exact opposite of the logic that determines higher growth companies deserve higher multiples? Why would high growth lead to high multiples at the micro level and low growth lead to them at the macro level?

 

If low GDP is the secret to high multiples, where are they in Germany and Japan 0?

[JimBowerman]

Isn't this the exact opposite of the logic that determines higher growth companies deserve higher multiples? Why would high growth lead to high multiples at the micro level and low growth lead to them at the macro level?

 

If low GDP is the secret to high multiples, where are they in Germany and Japan 0?

 

Buffett answers that question here:

 

[CorpRaider]

As I'm sure we all know, CAPE is at ~30; in 1999 it was at ~45; 10 year was at ~6%. 

 

So we could have room to run without setting any records (and without even needing to talk about Japan). 

 

I wouldn't like to have to place a high-conviction bet either way.

[SHDL]

As I'm sure we all know, CAPE is at ~30; in 1999 it was at ~45; 10 year was at ~6%. 

 

So we could have room to run without setting any records (and without even needing to talk about Japan). 

 

I wouldn't like to have to place a high-conviction bet either way.

 

Right, which is why I wouldn’t be too surprised if we had a final stage of euphoria before this cycle ends.

 

One thing to note though is that corporate profits as a share of GDP have been unusually elevated over the last 10 years or so, and the CAPE ratio (by only looking at earnings over the last 10 years) effectively assumes that this is the “new normal.”  That may be correct, but if it’s not then the market is already just about as expensive as it was in 1999.

[TwoCitiesCapital]

As I'm sure we all know, CAPE is at ~30; in 1999 it was at ~45; 10 year was at ~6%. 

 

So we could have room to run without setting any records (and without even needing to talk about Japan). 

 

I wouldn't like to have to place a high-conviction bet either way.

 

Right, which is why I wouldn’t be too surprised if we had a final stage of euphoria before this cycle ends.

 

One thing to note though is that corporate profits as a share of GDP have been unusually elevated over the last 10 years or so, and the CAPE ratio (by only looking at earnings over the last 10 years) effectively assumes that this is the “new normal.”  That may be correct, but if it’s not then the market is already just about as expensive as it was in 1999.

 

;) Margins are at a record and appear to be on the cusp of contracting given wage rates rising 3-4% and rising commodity/input costs - much higher than revenue growth/GDP.

 

Then again, markets didn't care in 2015 when earnings contracted for an entire year so maybe we don't care now.

[TwoCitiesCapital]

Isn't this the exact opposite of the logic that determines higher growth companies deserve higher multiples? Why would high growth lead to high multiples at the micro level and low growth lead to them at the macro level?

 

If low GDP is the secret to high multiples, where are they in Germany and Japan 0?

 

Buffett answers that question here:

 

 

1) there's a caveat to what he said which basically is summarized 'if you have the certainty rates will remain low.' Yes, if rates remain at 0%, stocks CAN trade at massive premiums into perpetuity. We don't have that certainty AND it doesn't mean stocks SHOULD trade at massive premiums

 

2) low rates make financing easier and act a sa grease on financial systems. This makes it EASIER to bid up assets which, in turn, causes financing to become even easier. It's a fortuitous cycle, but it says NOTHING of the intrinsic/fair value of the stock. Only that the environment breeds easier financing/speculation/and potentially higher values

 

3) using DCF, the value of a stock is the summation of all cash flows discounted into perpetuity. Basically it's  CF (1+growth)^n/(1+rates)^n

 

People who say that low rates lead higher values/multiples assume it's because your denominator is shrinking. But if those low rates are due to exceptionally low growth/low inflation, then your numerator is ALSO shrinking. Assuming higher multiples is double counting the benefit of low rates.

 

Example scenario. $1 invested at 6% growth for 30 years when discount rates are 5% has a present value of $34.86. A P/E of ~35x. 

 

If you assume the discount rate falls to 4% you get a P/E of 40 and a present value of $40; however, that doesn't consider the impact of the lowered growth. If the 1% drop is primarily due to a 1% drop in inflation or growth, then your numerator also shrinks and your stock's present value is virtually unchanged at $34.91. The same 35x.

 

People who say low rates lead to higher multiples are 1) confusing the meaning of CAN and SHOULD and/or 2) ignoring mathematics and/or 3) ignoring real life cases like Europe and Japan where the same relationship HASN'T held true

 

 

[scorpioncapital]

Some argue that Japan and Europe have policies that are anti-growth and deflationary therefore rates can be at 0 or negative and it wouldn't really work. In the end it becomes the policies that unleash the animal spirits and human potential that matter.

[CorpRaider]

As I'm sure we all know, CAPE is at ~30; in 1999 it was at ~45; 10 year was at ~6%. 

 

So we could have room to run without setting any records (and without even needing to talk about Japan). 

 

I wouldn't like to have to place a high-conviction bet either way.

 

Right, which is why I wouldn’t be too surprised if we had a final stage of euphoria before this cycle ends.

 

One thing to note though is that corporate profits as a share of GDP have been unusually elevated over the last 10 years or so, and the CAPE ratio (by only looking at earnings over the last 10 years) effectively assumes that this is the “new normal.”  That may be correct, but if it’s not then the market is already just about as expensive as it was in 1999.

 

I think I failed to follow. How would the share of profits to GDP and/or margins work into measure the dollar of market cap paid share of smoothed net earnings (right?).  So like market cap to GDP is equal to 99? 

 

Many would probably also argue that profits to GDP is higher in 2009 than 1999 because of greater intl trade/share of S&P 500 revenues coming from abroad (e.g., like the FANGS dominating industries across the globe).

 

I personally think margins will mean revert or at least trend that way, but I've (wrongly) thought for a number of years. 

[Cigarbutt]

As I'm sure we all know, CAPE is at ~30; in 1999 it was at ~45; 10 year was at ~6%. 

 

So we could have room to run without setting any records (and without even needing to talk about Japan). 

 

I wouldn't like to have to place a high-conviction bet either way.

 

Right, which is why I wouldn’t be too surprised if we had a final stage of euphoria before this cycle ends.

 

One thing to note though is that corporate profits as a share of GDP have been unusually elevated over the last 10 years or so, and the CAPE ratio (by only looking at earnings over the last 10 years) effectively assumes that this is the “new normal.”  That may be correct, but if it’s not then the market is already just about as expensive as it was in 1999.

 

I think I failed to follow. How would the share of profits to GDP and/or margins work into measure the dollar of market cap paid share of smoothed net earnings (right?).  So like market cap to GDP is equal to 99? 

 

Many would probably also argue that profits to GDP is higher in 2009 than 1999 because of greater intl trade/share of S&P 500 revenues coming from abroad (e.g., like the FANGS dominating industries across the globe).

 

I personally think margins will mean revert or at least trend that way, but I've (wrongly) thought for a number of years.

I don't want to get directly involved in this dicussion but I've read this in the past and thought the discussion fun to read and Mr. Montier talks about the margin effect:

http://lwmconsultants.com/wp-content/uploads/2014/04/JM_CAPECrusader1.pdf

 

 

[SHDL]

As I'm sure we all know, CAPE is at ~30; in 1999 it was at ~45; 10 year was at ~6%. 

 

So we could have room to run without setting any records (and without even needing to talk about Japan). 

 

I wouldn't like to have to place a high-conviction bet either way.

 

Right, which is why I wouldn’t be too surprised if we had a final stage of euphoria before this cycle ends.

 

One thing to note though is that corporate profits as a share of GDP have been unusually elevated over the last 10 years or so, and the CAPE ratio (by only looking at earnings over the last 10 years) effectively assumes that this is the “new normal.”  That may be correct, but if it’s not then the market is already just about as expensive as it was in 1999.

 

I think I failed to follow. How would the share of profits to GDP and/or margins work into measure the dollar of market cap paid share of smoothed net earnings (right?).  So like market cap to GDP is equal to 99? 

 

Many would probably also argue that profits to GDP is higher in 2009 than 1999 because of greater intl trade/share of S&P 500 revenues coming from abroad (e.g., like the FANGS dominating industries across the globe).

 

I personally think margins will mean revert or at least trend that way, but I've (wrongly) thought for a number of years.

 

That was pretty unclear, sorry.  The valuation metric I was referring to is basically a modified CAPE ratio.  Like you say, the idea is to measure market cap divided by normalized earnings.  The question is how to measure normalized earnings.  CAPE has its way of doing it, which is to basically look at average earnings over the last 10 years.  That’s a decent way of filtering out business cycles, but is it enough?  For instance, if we eyeball the following chart, we can see that the corporate profit to GDP ratio seems have gone through slow moving cycles, and was ~20% below its long term average during the 1990s and was ~20% above its long term average during the 2010s:

 

https://fred.stlouisfed.org/graph/?g=nw16

 

This suggests that maybe it makes sense to adjust for this “cyclicality” too — which CAPE doesn’t — by increasing the CAPE “E” for 1999 by 20% and reducing the CAPE “E” for 2019 by 20%.  If we do so, we get a similar “adjusted CAPE” for both 1999 and 2019.

 

And yes, you’re right that there are solid explanations for why the profit to GDP ratio has stayed elevated in recent years.  I share your skepticism as to whether things will stay that way forever.

[SHDL]

;) Margins are at a record and appear to be on the cusp of contracting given wage rates rising 3-4% and rising commodity/input costs - much higher than revenue growth/GDP.

 

Yes, that appears to be the case:

 

https://www.wsj.com/articles/investors-brace-for-hit-to-profits-as-costs-rise-11554283800?mod=hp_major_pos20

 

Now I’m actually starting to get a little worried/excited… 

 

First, we have these reports of rising labor/input costs, which are obvious negatives for corporate profits.  But it doesn’t stop there because that almost surely also means we are looking at higher inflation down the road, which gives the Fed a very good reason to start tightening again.  That, for one thing, means higher interest expenses and hence lower profits.  It also means we’ll likely have lower P/E multiples.  So for equity investors it’s a double whammy of lower profits & lower valuations.  And to matters worse, there is so much corporate debt outstanding, much of it apparently mis-rated, that even a modest reduction in profits could trigger a big spike in credit spreads — which may well amplify the effects above in an unpleasant way.