Progress of the secular bear market: position as of November 30, 2007
The value for R is 1500 as of November 2007. For S&P500 of about 1480 this gives P/R of 0.99.
Is the bull market over?
Nine years ago I wrote on my first stock market articles in which I asked a very similar question about the secular bull market that had begun in 1982. It was a time of financial turmoil with the start of the Asian crisis the year before and the Russian debt default in that year. All of this was topped off by the LTCM collapse and a near 20% drop in the stock market. It was a scary time, but my view was cautiously optimistic solely based on the fact that stock valuations (using my P/R tool) were too low for this to be the end.
A year later I had changed my view. The market had risen strongly and by summer 1999, it had reached a level consistent with the highest historical valuations. I wrote another article in fall 1999 that proclaimed the end of the secular bear market. I was early, the bull did not end until 2000. This month's article is going to make that same sort of argument that I made in 1998, that the bull has yet a bit more to run.
As in 1998, I have no watertight argument for why the current mortgage crisis will not produce a recession or bear market. My reasons for believing the bull will continue are based on historical comparisons. Firstly, the current expansion, at six years duration, is still to short compared to post-1960 business cycles. The 1960's business expansion was 8.83 years long. The one in the 1980's was 7.67 years long and the one after that was 9.75 years long. Only the 1970's expansions were shorter, but this was due to recessions induced by the oil shocks caused by the 1973 embargo and the 1979 Iranian Revolution. Baring another shock, I do not believe the 1970's business cycles are valid examples. Of course, at close to $100 a barrel oil sells for about what it did (in constant dollars) in the late 1970's. Could this not induce a recession like it did in the early 1980's? My answer is yes, but not quite yet, as I will get to later.
At present we do not have conditions like those around 1980 and there has been no constriction in the supply of oil, and so no "oil shock". And until we do have this, I believe that the length of this expansion will fall into the 7.7-9.7 year region, giving the timing for the start of the next recession in the 2009-2011 window. This timing suggests a bull market peak in the summer of next year at the earliest, implying more bull to go.
The second argument is based on P/R (see graph at top of article). In 1999, P/R reached levels higher than any before seen, which was why I made my premature call of the top in the fall of that year. The actual peak came in 2000. The bear market that followed was more extreme than any seen since the Depression. The graph shows how the initial bear market of the current secular bear market was greater than all the others shown in the figure expect for the one in the Depression. The bull market since 2002 is the weakest bull market of all the secular bear markets shown in the figure. After five years, P/R has managed to briefly reach 1.0, two thirds of its peak values in 1999-2000. This value is much below the ~1.2 level seen in the first bull market peak of previous non-Depression secular bear markets and lower than the levels seen at the second bull market peak. The second bull market peak corresponds better in time with the present because of long cycle length today. These peaks are about 80% of their initial peak P/R levels, and so in relative terms today's market level is even lower than those of the past.
In other words, the story told by P/R is much like the one told in 1998. Although the idea that the bull is over cannot be ruled out on the basis of P/R (see figure) P/R simply isn't very high. Its level is more consistent with a time in which "irrational exuberance" has been chastened. For example, the three most spectacular stock market "crashes" in the 20th century have been the 1929 crash, the 1987 crash and the dotcom meltdown. All three occurred during periods during which free market nostrums have been widely accepted and taxes (particularly those on capital gains) were low. In fact, it was the 1997 reduction in capital gains that convinced me that P/R would reach extreme values in the last secular bull market and so the bull market was probably not over in 1998. Back then, like now, I was fairly heavily invested, having not yet liquidated my 401(k) stock positions (I would do this in the next year).
Looking at the P/R figure it appears to me that there is still room for a final run up before the end in 1-2 years. Today the earnings yield on the S&P500 is about 5.3%, 1.3% higher than that on 10-year Treasuries. Back in 1998, the earnings yield was 4.1%, 0.6% lower than the yield in 10 year Treasuries. The risk premium today is nearly 2% greater than then.
Of course the world is different today. Today the specter of inflation has raised its head with an extraordinary rise in the price of commodities, particularly oil, which is reminiscent of the 1970's. Could a sharp increase in oil price act as a supply shock as it did in the 1970's? The short answer is yes. A very rapid rise in price would have the same impact as a supply constriction, forcing a reduction in economic activity, that is, a recession.
In 2005, I wrote an article[1] in which I expressed the opinion that high oil prices need not induce a recession:
This analysis suggests that high oil prices do not have to derail economic growth, allowing the current expansion to last the full Juglar cycle (i.e. until ~2008) as it has in three of the past four decades. Oil prices can stay high, Asian demand can continue to grow (barring financial crisis or recession there) and US interest rates can stay low. US oil consumption will fall in response to high prices, creating demand for more efficient vehicles and appliances, which will help support economic expansion. High oil prices should also stimulate interest in alternate energy technologies which could serve as additional leading sectors to enhance US economic growth.
The analysis was based on a plot of real oil price and the ratio of oil consumption to GDP over time. At around $45/barrel (annual average price in 2005 dollars--about $47 today), an acceleration in this long-term decline in this ratio was noted. I forecasted that once a running one year average of oil prices rose above this level, another increase the in the downward slope of the oil/GDP line would occur just as it did in the late 1970's. This change in slope would reflect substitution of higher efficiency vehicles for lower efficiency ones.
The ratio of oil to GDP is a measure of the importance of oil to the economy. This ratio grew until 1954, meaning that oil use was a leading sector of the US economy. It contributed more than its share to economic growth and the US economy was maximally sensitive to oil supply disruptions. Since 1954 the ratio has fallen, meaning the importance of oil to US economic growth has been falling. The sharp change in slope of this ratio meant that a price of $47/bbl (in today's dollars) was sufficient to spur either an increase in oil use efficiency or substitution of something else for oil, which resulted in a reduction in the overall amount of oil the US economy consumed.
Average oil prices did exceed $47 in 2005, and they were yet higher in 2006 and higher still in 2007. Although overall oil consumption stopped rising in 2005, it hasn't started to fall despite prices rising well above the $47 level. My analysis needs to be refined. Figure 1 shows a plot of oil use for transportation, expressed as barrels per driver per year. Also shown is oil price in 2007 dollars and crude oil cost expressed as dollars per driver per fortnight (two weeks). The unusual units for this measure were chose to make the cost scale similar to the price scale to make comparison easier.
This figure tells a different story that the crude analysis using total GDP and total oil use. An 18% drop in oil use per diver was seen in the ten years after the 1973 oil embargo. Some of this decline was the result of the 1982-83 recession however and doesn't count as an efficiency-based reduction. The 7.4% decline in 1973-75 certainly reflects the recession then. This decline is virtually the same as the 7.5% decline from 1980-1983. This implies that the 11% decline over 1973-80 reflected a real increase in transportation efficiency.
The 1980 oil consumption of 24.1 barrel/driver/year was not sustainably surpassed until after 1997, when oils prices had returned to their pre-embargo levels. Since then oil consumption has soared, reaching the levels in 2005 not seen since 1979. Not only that, but oil cost per driver has been at 1979 levels over the last two years, and if recent spikes in price to the $90 region persist for a year, oil cost will reach 1980's levels, the highest levels ever seen.
As the figure shows, oil consumption per driver has begun to moderate and in 2007 seems poised to come in at levels below those of the last three years. Thus it seems that a move towards efficiency has begun, but it is much weaker than in the 1970's. The explanation for this is straightforward. Efficiency improvements in the 1970's were mandated by the CAFE standards established by Congress in the wake of the oil embargo. Today it is telling that Congress is again discussing major increases in CAFÉ standards that if enacted would like move the oil use line downward again.
Figure 1. Trends in oil price and utilization for transportation 1960-present (data from refs 2-5)
There is another difference between today and the 1970's. There was a major reduction of total oil use in the late 1970's that did not reflect improvements in automobile efficiency. Between 1973 and 1983 total oil consumption for transportation actually rose 3.9%, despite the reduction in oil use per driver. The increase in drivers overwhelmed the improvement in efficiency. In contrast there was a 29% decrease in oil use for non-transportation purposes. In fact oil consumption today for non-transportation purposes is still 20% lower than it was in 1973. So the GDP analysis I made in my 2005 article was valid as far as the amount of oil needed to produce a given amount of GDP was concerned. High oil prices did cause a more or less permanent shifting in how oil is used in non-transportation sectors. That is, Business did respond to high prices by increasing efficiency or using substitutes as one would expect rational economic actors to do.
Consumers respond differently. Indeed, despite the availability if high mileage cars, consumers have continued to show a preference of low mileage trucks and SUVs even in the face of oil prices that caused substantial retrenchment three decades ago. It would appear than the reduction in oil use by consumers back in the late 1970's was caused by the nonavailability of fuel inefficient new cars caused by CAFE standards. Consumers cannot buy gas hogs if none are for sale.
Since then, as trucks designed as passenger vehicles have become available (trucks have lower efficiency standards than cars do) people have opted to buy them. Efficiency has declined and oil use has risen: average fuel efficiency for the total new car+truck fleet in 2001 was some 7% lower than the peak level achieved in 1986[6] and oil use per driver is up 6% over that same time (Figure 1).
My argument that oil consumption would start to decline if prices remained at their 2005 level or went higher was based on historical observations involving on overall oil consumption, much of which was used for non-transportation purposes. Oil used for transportation was largely price insensitive, the modest decline seen reflected government action, not a specific effect of high prices. Since transportation is the dominant use for oil today and there have been no government mandates to increase vehicle efficiency, no significant reduction in oil use would be expected and indeed none has appeared.
High oil price without an exogenous supply restriction can serve as an oil shock and induce an "early" recession all by itself. In fact, this may be what brought the early 1980's recessions, which ended an expansion at 4.8 years, much shorter than in other recent business cycles. What this means is if oil prices remain at the $90+ range for the next year, the real cost to motorists will match the situation in 1980 and recessionary conditions could develop in 2008, just as in 1980. And just like in 1980, the Fed will likely halt or even prevent recession by aggressively cutting rates. This will not stave off recession for long--recession recurred in the second half of 1981. Thus recession is likely in 2009 if this scenario plays out.
A recession in 2009 falls into the range specified above based on cycle length. So even if oil prices do stay high it is likely that the time table I implied back in February 2005[7] will still hold. I will note that 1980 was an excellent year for the market with 30%+ gains, and I expect 2008 to be a good year as well. But it looks increasingly likely that it will be the last good year for quite some time.
References:
[1] Alexander, Michael A., "Oil Prices, the Kondratiev Cycle and Peak Oil", Safehaven, April 22, 2006.
[2] WTI price and CPI comes from Economagic.com
[3] Number of drivers comes from Wikipedia article, "Passenger vehicles in the United States": http://en.wikipedia.org/wiki/Passenger_vehicles_in_the_United_States#Total_number_of_vehicles
[4] Crude oil consumption from Annual Energy Review, Energy Information Administration, "Table 5.13c Estimated Petroleum Consumption: Transportation Sector, 1949-2006"
[5] 2007 Crude oil use estimate from High Beam Encyclopedia: (www.encyclopedia.com/doc/1G1-171581264.html)
[6] Kathy Gill, "CAFE Mileage Standards", About.com:politics (http://uspolitics.about.com/od/energy/l/bl_cafe_table.htm)
[7] Alexander, Michael A., "Charting the Course of the Secular Bear Market", Safehaven, February 21, 2005.
