The Twenty Year Stock Bubble Is Still Inflated

If you talk to someone who is in their early 40s, they are likely to tell you that they have seen two bull markets and two bear markets in their adult lives. Because their experience has conditioned them to expect that cycle to repeat, many of this vintage will tell you it is time for another bull market. In fact, older investors with much more experience may have a completely different perspective. For someone with 40 or 60 years of market experience, there may be a different possibility considered. Based on the longer time frames, the possibility of a long term bubble may be recognized.

Barry Ritholtz at The Big Picture (here) has a telling graph, courtesy of Ron Griess at The Chart Store.

If this doesn't give you pause to think, then get a reading from the numbness meter. If we were to return to the historical average, GDP would have to rise by 67% with current stock prices or stock prices would have to fall by 40% with current GDP. Here are some examples for stock price decline at constant GDP that would get to the average:

a. 5% average annual decline rate for 6.9 years.

b. 4% average annual decline rate for 8.6 years.

c. 3% average annual decline rate for 11.5 years.

If stock prices remain constant, here are examples of GDP growth that would get to the historical average:

a. 4% average annual growth for 13 years.

b. 3% average annual growth for 17.4 years.

c. 2.5% average annual growth for 20.7 years.

If GDP were to grow at an average annual rate of 2.5% for five years, the stock market return to get to the historical average would be -32.2% total return, or -5.7% average annual return.

If GDP were to grow at an average annual rate of 3% for five years, the stock market return to get to the historical average would be -30.4% total return, or an average annual return of -5.5%.

If GDP were to grow at an average annual rate of 4% for five years, the stock market return to get to the historical average would be -27% total return, or an average annual return of -4.9%.

There will be those who dismiss long term historical relationships between value and stock prices. They do that at their peril. If something is priced above its value it can stay there a long time, but likely not forever. Just as many wise men did not perceive the Japanese bubble, the dot.com bubble and the worldwide credit bubble, there will be wise men who do not see other bubbles.

We are still in a bubble in stocks, now twenty years in duration. This bubble may not deflate to or below historical averages for many more years or it may do so within one or two years or less. Many wise men will be surprised again. Don't be one of them.

Comments

[Dr. O]

What about wise women?

Hard to say, what with 0% interest rates, and high commodity prices. It's like we're in the midst of recession-depression for most people, but speculators have high access to free money. Credit for most people is tough to get, and small businesses are crumbling.

Ultimately, I think stocks will go back down before economic activity picks up enough to support asset prices.

Personally, I'm about hard assets.

Long GLD, TWM, SDS.

2009 Nov 20 06:53 AM

[OstrichHater]

This is a very interesting analysis, but the author's calculations seem to miss the impact of inflation on nominal GDP growth. I believe the authors forecasts are likely for real GDP growth, which isnt the whole picture.

With the massive amount of money printing that the Fed has done and will likely continue to do, we are more than likely going to get inflation for the next 5-6 years. I suspect that inflation will ramp up to 4% over the next couple of years, possibly even higher. That combined with GDP rebounding immediately in late 2009 and early 2010 (~5%), and then tapering off to a 2-3% range, its not unlikely that we will have nominal GDP growth of greater than 6%, even while real GDP growth lags historical trend rates.

This could bring the relationship back much faster (5-6 years) with no or very limited stock market appreciation.

2009 Nov 20 07:27 AM

[Andrew Butter]

John

If that chart is supposed to represent a valuation based on GDP (a measure of aggregate income and/or expenditure), shouldn't there there be a yield somewhere in the equation to be dimensionally correct?

If you allow for the cost of money as some sort of a function of long-term interest rates you get a different result which is intuitively more in line with the idea that the stock market in 1929 was more over priced (more of a bubble) than now.

2009 Nov 20 07:34 AM

[a fat panda]

Great article. I would like to have seen you discuss profit margins as well. Here is why: I don't dismiss the historical relationships, but I would ask you how do you factor in the advent of the PC? You will note the break with the past started in early 90s when PCs were just gaining wide acceptance in the market place. Adding the PC made 1 do the work of two. In my case, I went from programming on 3 PCs to just 1 because the speeds were faster. Our contribution to GNP didn't increase all that much, but we did it with a fraction of the input costs. This is about increasing profit margins which made the value of the business go up a lot.

2009 Nov 20 07:44 AM

[BSD77]

Interestingly, if you compare the S&P 500 growth and US GDP growth over the 1975-2009 period, you come to a pretty much the same conclusion. In order for the S&P 500 to reflect the (nominal) GDP growth over that period the S&P should be around 700 points today.

2009 Nov 20 07:44 AM

[enigmaman]

If you look back at the charts for DJ & S&P& NDQ since 1990 like it was a mutual fund or a stock that was available to buy, would you buy it? Its that simple, there should be no difference at all, just like you would consider buying Berkshire Hathaway shares. What you would use to do fundamental analysis is the US economy, so would you buy it? If so Why? The current economic path we find our selves on is counter everything that would be needed to justify buying based on fundamental analysis, so why are people still buying? Is it that they see a long bright future for America, are they buying for short term gain without concern for the future or do they just close there eyes to reality because they dont like what they see and buy hoping it will be better tomorrow. In any case none bode well for our future and the future of the security in question, that being America!

2009 Nov 20 07:50 AM

[enigmaman]

Or worse!


On Nov 20 07:27 AM OstrichHater wrote:

> This is a very interesting analysis, but the author's calculations
> seem to miss the impact of inflation on nominal GDP growth. I believe
> the authors forecasts are likely for real GDP growth, which isnt
> the whole picture.
>
> With the massive amount of money printing that the Fed has done and
> will likely continue to do, we are more than likely going to get
> inflation for the next 5-6 years. I suspect that inflation will
> ramp up to 4% over the next couple of years, possibly even higher.
> That combined with GDP rebounding immediately in late 2009 and early
> 2010 (~5%), and then tapering off to a 2-3% range, its not unlikely
> that we will have nominal GDP growth of greater than 6%, even while
> real GDP growth lags historical trend rates.
>
> This could bring the relationship back much faster (5-6 years) with
> no or very limited stock market appreciation.

2009 Nov 20 07:51 AM

[cyclingscholar]

Mr. Lounsbury, 'market capitalization' is a function of how many shares are publicly traded, as well as the $ Level of those shares. As more companies went public in the 1980s and 1990s, i think this biases your data upward. A bubble is more likely to be detected using valuation measures like P/E, P/Cash flow, P/Sales, etc. If the NYSE and NasDaq tightened listing restrictions (heavens!) your plotted line would collapse.

2009 Nov 20 07:54 AM

[Flav]

So, have you test the stationarity of the series in order to claim that a historical average has any meaning? In case you're unaware of, i remind you that when a time series is not stationary, then both its mean and its variance depend and hence change with time, so to compute a historical mean is absolutely worhtless, as i think it is the case here.
Please, before making any statement regarding a financial time series, at least try to make a Dickey-Fuller test

2009 Nov 20 08:01 AM

[enigmaman]

Yes and "If the cat were a chicken, she'd lay eggs"


On Nov 20 07:54 AM cyclingscholar wrote:

> Mr. Lounsbury, 'market capitalization' is a function of how many
> shares are publicly traded, as well as the $ Level of those shares.
> As more companies went public in the 1980s and 1990s, i think this
> biases your data upward. A bubble is more likely to be detected
> using valuation measures like P/E, P/Cash flow, P/Sales, etc. If
> the NYSE and NasDaq tightened listing restrictions (heavens!) your
> plotted line would collapse.

2009 Nov 20 08:02 AM

[Dave Wrixon]

I think the obvious flaw in the argument is that things don't return to long-term averages. Most correction tend to overshoot by a similar magnitude to the original excess. This tends to suggest that your prognosis is hopelessly over optimistic, even ignoring the negative effects of the corruption and myopic meddling that has been taking place. Furthermore, the analysis takes no account of the fact there is a fundamental and irreversible global power-shift taking place that can only have the effect of removing the US from its pedestal of privilege. Of course there is possibility that this inevitable shift combined with a cyclical correction of extreme magnitude could result in a cataclysmic economic event.

And I guess that I have just been dated.

2009 Nov 20 08:09 AM

[TheFounder]

It can also be argued that since the market is the sum of all the wisdom of all the wise men (and even better - the wise women), then at current levels, the wise men and women believe that inflation will be at 8-12% for 5-7 years.

2009 Nov 20 08:11 AM

[BSD77]

I see that sbd. was sitting for the Level 2 exam ;-) but still I'm not sure that Dickey Fuller test is relevant here - it is a test for non-stationarity of data that's for sure but it would be more appropriate for a time series of one dependent variable in time, not a ratio, since both of the variables here are (or should be) strongly correlated and hence there ratio should be constant over time.


On Nov 20 08:01 AM Flav wrote:

> So, have you test the stationarity of the series in order to claim
> that a historical average has any meaning? In case you're unaware
> of, i remind you that when a time series is not stationary, then
> both its mean and its variance depend and hence change with time,
> so to compute a historical mean is absolutely worhtless, as i think
> it is the case here.
> Please, before making any statement regarding a financial time series,
> at least try to make a Dickey-Fuller test

2009 Nov 20 08:14 AM

[chap08]

What happens if you consider Enterprise Value, rather than Capitalization?

It is not just consumers and government that have increased debt over recent years. Corporates have too - big time. This means that Capitalization can grow rapidly during the goods times, due to the multiplier effect of leverage on earnings. P/E ratios can look still reasonable, while earnings / enterprise value starts to look bad.

Obviously the problems come during the bad times. Now, capitalization has to fall in a leveraged way, just as it rose in a leveraged way during the good times. This means that if there is going to be future earnings volatility (seems likely) then the graph understates the current level of over-valuation.

2009 Nov 20 08:19 AM

[bartpr]

your analysis follows the same conclusion as the pe10, the spy price to 10 years average earnings. look at each. they are correlated.

2009 Nov 20 08:25 AM

[BSD77]

Nope, they believe that nominal GDP growth will be in that range (5y 11% or 7,6% 7 years). So, if we use a historical long- term real GDP growth (app. 3%) as a good measure for these next couple of years, they believe that inflation will be 4,6% to 8% in that period. Unless, ofcourse, the stock market continues it's way down.


On Nov 20 08:11 AM TheFounder wrote:

> It can also be argued that since the market is the sum of all the
> wisdom of all the wise men (and even better - the wise women), then
> at current levels, the wise men and women believe that inflation
> will be at 8-12% for 5-7 years.

2009 Nov 20 08:26 AM

[bbro]

Dont you think you are operating in a bubble by not accounting for the
interest rate environment....1973 to 1990 had a much higher inflation
induced environment....as well 1996 to 2001 was tech crazed....
If you want to look at valuations look at NIPA corporate profits to S&P 500
in a low Inflation environment ( under 4% year over year) you will
see a drastically different picture. On the same token I am glad to see
articles like this because it porvides the wall of worry which most markets need to go up. Sir John Templeton says bull markets end in extreme optimism.....

2009 Nov 20 09:01 AM

[bbro]

Sir John Templeton said....

2009 Nov 20 09:05 AM

[Flav]

But market capitalization and GDP are both non-stationary series, so, why would you expect a variable constructed as a ratio of this two to be stationary? I believe the Dickey-Fuller test is still suitable for this situation, as you define the new variable as MC/GDP


On Nov 20 08:14 AM BSD77 wrote:

> I see that sbd. was sitting for the Level 2 exam ;-) but still I'm
> not sure that Dickey Fuller test is relevant here - it is a test
> for non-stationarity of data that's for sure but it would be more
> appropriate for a time series of one dependent variable in time,
> not a ratio, since both of the variables here are (or should be)
> strongly correlated and hence there ratio should be constant over
> time.

2009 Nov 20 09:11 AM

[enigmaman]

John Templeton was correct concerning a bona fide "Bull market" is that what this is or is it a rally in a long term bear market?

On Nov 20 09:01 AM bbro wrote:

> Dont you think you are operating in a bubble by not accounting for
> the
> interest rate environment....1973 to 1990 had a much higher inflation
>
> induced environment....as well 1996 to 2001 was tech crazed....<br/>If
> you want to look at valuations look at NIPA corporate profits to
> S&amp;P 500
> in a low Inflation environment ( under 4% year over year) you will
>
> see a drastically different picture. On the same token I am glad
> to see
> articles like this because it porvides the wall of worry which most
> markets need to go up. Sir John Templeton says bull markets end in
> extreme optimism.....

2009 Nov 20 09:12 AM