Could the Dow Fall to 1600?

The Dow at 1600 seems a little too scary. Can we really get that bad and fall 3/4 from this point? Hopefully we won’t. But economic theories don’t preclude that possibility. This is because of the fundamental relationship between the stock market and the economy.

Businesses produce profits and the profits impact both the GDP and stock values. You can have the profits go closer to zero and the GDP still at non-zero due to the other components in the GDP, but you cannot have the GDP going to a toilet while profits keep going up indefinitely. Thus, GDP forms an upper bound for corporate profits. Generally, the market capitalization of the entire economy is a product of GDP, corporate profits as a percentage of GDP and what investors think the future will be (Price to Earnings ratio).

Stock market values depends on three main factors in the short run -

1. Market psychology (affects Price to Earnings)
2. Growth in corporate profits (affects Earnings)
3. Interest rates (affects Price to Earnings)

However, as Buffet once said, “In the short run, the market is a voting machine; in the long run, it is a weighing machine”. In the long run, market psychology and interest rate growth don’t have much impact on stock valuation and corporate profits cannot grow forever – limited by GDP level. So, in the long run stock values would be linear in Gross Domestic Product (GDP) – that indicator of what the economy actually produces.

Figure 1 has the US GDP for the last 60 years in 2000 dollars. GDP has grown about 7 times since 1947. So you expect the stock market value to be somewhere in that region, right? The relationship held 'till about the early 1980s and then it went out of whack (Figure 2). It grew nearly 30 times since 1950. From 1950 to 1982, the Dow grew about 3.5 times while GDP grew about 3 times; they roughly tracked each other. However, since 1982, the Dow has grown 8 times (assuming a value near 6500), while the US GDP has grown only about twice. So, if the historic GDP relationship matters, Dow must come to a level of 1600 (twice what it was 27 years ago).

Some of the factors that partly helped this included an increase in foreign income (that gets to stock value but not to GDP), but for most part we formed a massive bubble built in the back of market psychology and historically low interest rates, while the corporate profits a a percentage of GDP increased due to the relaxation of rules. These cannot continue forever. We might be in for a painful readjustment towards long term averages and that could make the Dow at 2000 a reality.

US GDP for the last 60 years in 2000 dollars

Data Source : St. Louis Fed

Dow Jones Industrial Average (Blue) and S&P 500 index since 1950.

Source: Yahoo Finance

Comments

[investor88:]

Balagi has put forward powerful arguments based on relationship to GDP growth that Dow should be 1600 at the end of it all. If we take note that in severe markets like in 1929 and in many other markets the markets dropped 90pc from the peak, we see that 10% Dow 14k=1,400 or thereabouts is a possibility. Dow 1600 is of course not a certainty.

What is so ridiculous about Dow 1,600? Robert Pretcher holds the same view and many people ridicule him but who knows we just simply rule out the possibility.

Mar 03 08:13 AM

[boats:]

The larger question is when will rather than will the.

Mar 03 08:16 AM

[J. D. Swampfox:]

Buffet was quoting Benjamin Graham with the "weighing" quote. Graham argued the place to be when GDP, earnings and thus equity prices where on the decline was bonds and preferred stock, particularly of utilities. Back then, utilities were government regulated (controlled) monopolies. The government controls financials now and is really afraid to give the bondholders of financials a "haircut", maybe that's the new utility....?

(see: lockstep-investing.blo...)

Mar 03 10:04 AM

[Conan the B...:]

Robert Prechter predicted DOW 400 back in 1987, he had his head handed to him.

But the Double Top in the S&P worries me the most.

A 90% drop to 1600 on the DOW is one thing, the Move down by the S&P suggests Zero as a goal.

Thank Goodness 10% drops on a monthly basis for the rest of the year get smaller and smaller as the overall value decreases.

Mar 03 10:08 AM

[cabaretewil...:]

Time to buy some condos and villas in the emerging economy of the Dominican Republic. caribbeanrealty.ca Construction here is still booming, though I would expect a short slowdown in the next year. there are 86 million Babyboomers starting to retire right now. Enough of them have money to move south, and prices in the DR are at 50% relative to other islands. Check it out.

Mar 03 10:11 AM

[Le French:]

I think you made a good point by looking at the corrolation between GDP growth and price-to-earnings / valuation growth. However, you didn't take into account the devaluation of the dollar. Same as the commodity bubble, the equity markets have been fuelled in part by the weak dollar in the last few years. The price of your stock can go up but if the overall value of your investment goes down proportionally then you barely break-even. As a foreign investor, to make money buying US equities I had to outperform the market. The US dollar is mainly dependent on 3 factors:
(1) the level of US debt
(2) the US trade deficit
(3) Fed interest rates

The reason stock market has been like a slow motion train wreck and didn't crash is simply because of the recent political decisions which have strenghten the US dollar by leveraging temporarily consumer purchasing power and overvalued toxic assets, besides increasing rates of savings from people getting out of the capital markets in cash.

taken from an economic survey on crude oil prices:

"It is also of interest to note that the average nominal price for the OPEC basket of $24.36/B for the year 2002 was equivalent to only $6.53/B in 1973 dollars. The nominal price was more than eight times as much while the real price was only just over double. The main culprit was inflation and not the dollar."

source:
www.mees.com/postedart...

Mar 03 10:16 AM

[kelm:]

Good article. certainly if we look at the net fall in 1929 - 1932 1600 is a no brainer. The GDP analysis is good as well. The difficulty of course is the timing. This market has a nasty habit of moving decidedly down and then just hanging out and drifting around driving bulls and bears and buy and holders and day traders all crazy at the same time.

Mar 03 10:33 AM

[timhope:]

I think you're forgetting that GDP is inflation adjusted while your index returns are not. Since 1970 alone, we've had cumulative inflation of 560%. You should rebase them both to 1950 dollars and re-run the comparison.

Mar 03 10:47 AM

[mgcolin:]

I agree, seems like you're using a real GDP number. The inflation adustment would put the Dow much higher than 1600...use nominal GDP and to me the Dow is valued about right.

Also don't throw away your point on foreign earnings, many of the Dow companies derive 40-50% even more of their earnings overseas.

I do however think we'll see an overshoot on the downside and S&P 400 before this is all over.


On Mar 03 10:47 AM timhope wrote:

> I think you're forgetting that GDP is inflation adjusted while your
> index returns are not. Since 1970 alone, we've had cumulative inflation
> of 560%. You should rebase them both to 1950 dollars and re-run the
> comparison.

Mar 03 12:08 PM

[Robert H. H...:]

timhope is right. You're using the real GDP in chained-2000 dollars. Here's the link to nominal GDP.

research.stlouisfed.or...

Nominal GDP grew 53.4x from 1949 through 2008. In 1949 GDP was $267.3 billion. In 2008, it was $14,264.6 billion. That's a 7.0% compound annual increase.

Real GDP has grown 3.4% over the same period, a 7.1x increase.

Inflation accounts for the difference, having grown 3.8% per annum over the same period (This is CPI-all urban. PCE deflator data does not go back to 1950).

The fact that the DJIA tracked GDP prior to the 80's and then diverged probably has more to do with monetary policy in the two periods than anything else.

A few more cautions about this sort of analysis.

There are a number of disconnects between the level of a stock market index and the size of its home economy.

1) You mention the influence of international trade. That is surely important over a time series when companies become more global, earning profits overseas that do not show up in home country GDP.

2) Using a stock market index for this analysis will almost surely mislead.

It makes some theoretical sense that the total VALUE of traded stocks might have some constant relationship to the size of the economy over time, but you should look at the market capitalization of the companies in the index, not the index itself (subject to the following additional qualifications)

3) The DJIA is a particularly poor index to use for this sort of analysis as it consists of only 30 stocks, and is price-weighted rather than market cap weighted. Far better to use the S&P 500 or the Wilshire 5000.

4) Even adjusted for market capitalization, stock market indices (especially the DJIA with only 30 stocks) may exhibit "survivor bias" over long periods in the sense that poorly performing companies whose continued inclusion would pull the averages down are generally replaced by companies with brighter growth prospects.

5) The total value of publicly traded stocks (vs. privately held businesses) can change over time, reflecting the relative attractiveness of being public vs. private or the development of advanced markets. So for example, In 1950, Ford Motor Company was still a private company... It did not go public until 1956. So while there is probably a stable relationship between the total value of all businesses, -- public and private -- and GDP, you may be missing a variable if you look only at public values vs. GDP.

6) Corporate profitability can vary substantially over long observation periods, especially if there are changes in the tax burden falling on corporations vs. individuals.

7) Dividend policy can change dramatically over long periods, usually in response to tax rates on dividends vs. capital gains. Unless your index is rebased for dividends, significant changes in dividend payout ratios will lead to tracking error between the index and the profitability of the companies within it.

I'll stop here.

Even with all the disclaimers above, I'll re-run the analysis using nominal (not real) GDP.

The Dow started 1950 around 200. As mentioned above, GDP has increased to 53.4x the 1949 level. 53.4x 200 = 10,680.

That's not my price target for the DOW and please don't misread this as a bullish recommendation. But I'm not to worried about seeing 1,600 on the Dow soon.




On Mar 03 10:47 AM timhope wrote:

> I think you're forgetting that GDP is inflation adjusted while your
> index returns are not. Since 1970 alone, we've had cumulative inflation
> of 560%. You should rebase them both to 1950 dollars and re-run
> the comparison.

Mar 03 12:53 PM

[Chancer:]

That will certainly prove that indexing is the best- when you can buy the index for 1600.

Mar 03 01:13 PM

[henarl:]

Yes, the Dow could get down to 1600 in, say, 20 years but 20 years is a long time and lots of other things could happen in that time, some of which could make the Dow go even lower or maybe a lot higher than it is now. This kind of conjecture is more apropos to gypsies than market analysts

Mar 03 04:58 PM

[Balaji Visw...:]

Those are very good points Robert. I didnt account for inflation because the inflation roughly cancels the dividends from the stocks. The normal dividend yield is between 3 to 8% and that more than accounts for the inflation. So, my reasoning was that the growth in market cap after reducing dividends is more directly comparable with the growth in chained GDP value.



On Mar 03 12:53 PM Robert H. Heath wrote:

> timhope is right. You're using the real GDP in chained-2000 dollars.
> Here's the link to nominal GDP.
>
> research.stlouisfed.or...
>
> Nominal GDP grew 53.4x from 1949 through 2008. In 1949 GDP was $267.3
> billion. In 2008, it was $14,264.6 billion. That's a 7.0% compound
> annual increase.
>
> Real GDP has grown 3.4% over the same period, a 7.1x increase.<br/>
>
> Inflation accounts for the difference, having grown 3.8% per annum
> over the same period (This is CPI-all urban. PCE deflator data does
> not go back to 1950).
>
> The fact that the DJIA tracked GDP prior to the 80's and then diverged
> probably has more to do with monetary policy in the two periods than
> anything else.
>
> A few more cautions about this sort of analysis.
>
> There are a number of disconnects between the level of a stock market
> index and the size of its home economy.
>
> 1) You mention the influence of international trade. That is surely
> important over a time series when companies become more global, earning
> profits overseas that do not show up in home country GDP.
>
> 2) Using a stock market index for this analysis will almost surely
> mislead.
>
> It makes some theoretical sense that the total VALUE of traded stocks
> might have some constant relationship to the size of the economy
> over time, but you should look at the market capitalization of the
> companies in the index, not the index itself (subject to the following
> additional qualifications)
>
> 3) The DJIA is a particularly poor index to use for this sort of
> analysis as it consists of only 30 stocks, and is price-weighted
> rather than market cap weighted. Far better to use the S&amp;P 500
> or the Wilshire 5000.
>
> 4) Even adjusted for market capitalization, stock market indices
> (especially the DJIA with only 30 stocks) may exhibit "survivor bias"
> over long periods in the sense that poorly performing companies whose
> continued inclusion would pull the averages down are generally replaced
> by companies with brighter growth prospects.
>
> 5) The total value of publicly traded stocks (vs. privately held
> businesses) can change over time, reflecting the relative attractiveness
> of being public vs. private or the development of advanced markets.
> So for example, In 1950, Ford Motor Company was still a private company...
> It did not go public until 1956. So while there is probably a stable
> relationship between the total value of all businesses, -- public
> and private -- and GDP, you may be missing a variable if you look
> only at public values vs. GDP.
>
> 6) Corporate profitability can vary substantially over long observation
> periods, especially if there are changes in the tax burden falling
> on corporations vs. individuals.
>
> 7) Dividend policy can change dramatically over long periods, usually
> in response to tax rates on dividends vs. capital gains. Unless your
> index is rebased for dividends, significant changes in dividend payout
> ratios will lead to tracking error between the index and the profitability
> of the companies within it.
>
> I'll stop here.
>
> Even with all the disclaimers above, I'll re-run the analysis using
> nominal (not real) GDP.
>
> The Dow started 1950 around 200. As mentioned above, GDP has increased
> to 53.4x the 1949 level. 53.4x 200 = 10,680.
>
> That's not my price target for the DOW and please don't misread this
> as a bullish recommendation. But I'm not to worried about seeing
> 1,600 on the Dow soon.
>
>

Mar 03 06:55 PM

[San_Fran_Sam:]

I have just one word for all of you: Nikkei

the Nikkei is now back down to the level it was at in 1982!

and it took about 20 years.

so Dow below 1000 as it was in 1982 is not out of the realm of possibility.

Mar 03 11:44 PM

[Moon Kil Woong:]

Dow 1600 is not a possibility without a blow-up of the treasury note market/ US currency. In this case though, it would cause mass inflation as the dollar becomes worthless. A Dow 1600 would tend to mean GDP has contracted so much that the US debt burden is unsustainable, so I generally would rule out this scenario.

Mar 04 12:07 AM

[Conan the B...:]

Paul&Shark, Investor88 and kelm:

GDP inherently includes inflation, there is no need to adjust it to anything. Its a Chart of what it depicts, no more, no less.

The only misgiving I have with it, is the inclusion of 8 years worth of growth using 2000 USD for 2001 to present.

We are using a 90% decline based on the 1920's Bubble Burst, the Chart above for the DOW and S&P skips that period entirely.

On a personal Note: Connecting the Dots for both, I come up with the beginning of the Bull in 1982 and the end in 2002, roughly 20 years of ascendency.

Going back to the lows of 1974 and connecting thru the point in 1982, for the DOW, I get the 1K% mark, ( what that is relative to I do not have a clue since the Label says "Oct. 1, 1999" but the chart is extended to the present) . Doing the same for the S&P, I get roughly where we are right now.

The S&P is the more troubling of the 2 because of the Double Top. If there were no Labels whatsoever, I would get the same results and would use the sides of the Graph for measuring.

IMHO

Mar 04 12:43 AM

[Robert H. H...:]

Balaji Viswanathan -

Thanks for the response. I understand the "shortcut" you took with regard to dividends and the inflation rate, but I think it's led you astray.

It is true that the average dividend YIELD on the S&P 500 has been in the neighborhood of 3% in the post-war period (coincidentally similar to the inflation rate), but what is relevant to your analysis is the dividend GROWTH rate.

A simple version of the model you have in mind would assert that the following are (or should be) roughly constant over time.

1) Corporate Earnings : GDP
2) Equity Value : Corporate Earnings (P/E ratio)
3) Dividends / Earnings (Payout ratio)

For completeness, I might add the postulate that reinvestment returns are constant.

If true, then earnings will grow at the same nominal rate as GDP, and so will dividends. According to Hussman (2006 post), "The actual growth rate for S&P 500 dividends since 1940 has averaged just 5.7% annually."

Link: hussmanfunds.com/wmc/w...

If growth in dividends is proportional to growth in earnings, which is itself proportional to growth in the overall economy (all in nominal terms), AND market cap to earnings is a constant, then market cap will grow at the same rate as nominal GDP.

Mar 04 02:37 PM

[Conan the B...:]

Robert: yes, indeed.

Come friday the S&P's composition will change again, another realignment or out with the Low and in with the more expensive.

The components of each Index over the period of time described have changed radically. There is nothing Constant about either, especially on the dividend front.

Mar 05 12:47 AM

[reechurdjoh...:]

It sounds like fuzzy numbers to me. But if we have a severe depression, then 1600 could be possible. If you gave this number to Vegas, however, they would probably gladly give you a no limit bet that this will not happen.

Mar 05 09:41 PM