The massive fall of Treasury bill, note, and bond rates this past week reminds me of the remarks that Alan Greenspan made several years back about the bond "conundrum". Only more so. With the inflation we'd seen up to mid year 2008, interest rates should be far higher seven to nine years after the low of gold, and certainly not lower.
The retest of the rates lows (and the bond high) in September was understandable and short-lived. This past Thursday's event "feels like" a short squeeze of some sort, but who would have been short enough Treasury bonds to enable that? Or who is buying enough?
One thought I've seen in several places in the past few days is that if deflation seemed probable, the FED might feel the need to buy Treasuries. I assume that means if lowering FED funds to zero doesn't work to stimulate loan demand, and if the Keynesian "liquidity gap" has therefore gaped wide open, then they'd quickly lower the long bond rate. Here is Lyle Gramley's version of the story:
""Credit availability certainly hasn't increased," said Lyle Gramley, a former Fed governor. "That has to be a major concern for the Fed because historically the way we get out of recessions is having the Fed push down hard on the accelerator. If that is not working very well, we have to look somewhere else for salvation."
When the FED buys Treasuries it is "monetizing the debt", as it is usually called. The FED has already exchanged Treasuries to the banks for bad bank debt as part of the "re-financing". If they are buying Treasuries back they are cashing the banks (and others) out of their new (and old?) Treasuries and causing a short squeeze? This puts the favored banks flush with cash and the FED flush with bad debts.
The Treasury itself has recently sold scads of new bills, notes and bonds and taken cash *out* of the economy (somewhere in the world economy) for its on-again, off-again "rescues". So the FED monetization would seem to be a "reverse sterilization" to put the demand potential cash *back into* the economy that the Treasury has pulled out. Shell game or Princeton economics graduate seminar on solutions for the liquidity gap?
Many market analysts and even financial journalists are noting the fact that total returns of major stock indices (including all dividends reinvested and before any taxes) have under-performed US T Bills for the past ten years. (Right click on chart for larger version.)
Even international stocks with their great increases from 2003 to 2007, represented by Vanguard's International Value Fund, have underperformed the Vanguard T Bill Fund since November 1998, nor have small cap stocks done meaningfully better. The S&P500 Index has a total return of 1% per year. (Note: on this FastTrack charts BP = total ten year return and Ann = annualized total return for the ten years.) The two best ten year returns in this group were those of the Total Bond fund (+ 4.97% annually) and the Vanguard Long term Treasury Bond Fund (+ 6.33% annually). Mull those numbers over a bit and the fact that stocks did worse, on average.
One immediate question that rushes to mind and lips is "OK, stocks turned out badly, but did anyone predict it in advance?" Well, yes, a number of people usually referred to as value analysts, advisors, or investors did predict that stocks would under-performed T Bills. One person whom I remember saying so at the time is John Hussman--I didn't believe him then--and whose work is still available on his website as originally written in 1999. http://www.hussman.net/html/peak2pk.htm
"...From an analytical perspective, nothing affects the stock market except by affecting these three factors: dividend payouts, earnings growth, and P/E ratios. The primary effect of interest rates is on the P/E ratio. Falling interest rates encourage falling earnings yields (higher P/E ratios). Rising interest rates encourage rising earnings yields (lower P/E ratios). These effects are strongest when P/E ratios begin at extreme levels.
"If you understand this, you also understand why the market faces extreme danger now, and perhaps for several years into the future:
"1) Returns due to dividend income have never been lower...
"2) Earnings are near the peak of their long term growth channel...
"3) Stock prices have never traded at a higher multiple of peak earnings...
"Add these factors together, and investors face a long term total return of 7% annually if P/E multiples remain fixed at record highs, and earnings grow along the peak of their long-term growth channel. If the P/E contracts toward normal levels instead, stocks may very well under-perform Treasury bills for more than a decade (as they did from 1965 through 1986)."
There you have it. Identification of the factors which most affect the stock market, evidence that those three factors were then all at historically unsustainable extremes, and what the end result would be. The first time I recall seeing Hussman Funds' chart of 10 year projected stock prices based upon the current peak market P/E was in this 2005 chart (right click to open in new window):
This is where the chart was shown, and this was the conclusion:
" At the market's actual 2000 peak, valuations were so high that even a future price/peak earnings ratio of 20 could have been expected to result in a nearly zero annualized returns over the following 10 years. Not surprisingly, in the 5 years since the 2000 market peak, the S&P 500 has actually produced a total return of about –2% annually. The likelihood is that the coming 5 years will not be substantially better.
"Presently, the likely range of S&P 500 annual total returns for the coming decade is in the 2-3% range based on average and median scenarios, with outside possibilities as low as -3% in the very bearish case and still less than 8% in the very bullish case. "
Another thought which rushes to many minds about all this is, "OK, stocks indexes were a lousy investment, but I traded and invested in hot sectors and I made a lot of money from 2002 to 2008." And so did I, but higher risks were there in general and in those sectors. As a believer and practitioner of the economic long wave theory, and a former commodity futures trader, I felt certain that inflation and global growth were ready to go up by 2001, and I was heavily into energy and commodity and foreign stocks. However, most people and most institutions are not traders nor should they be. Most of them are investing to meet far distant goals for themselves or others: retirement funds, and the funding of colleges and universities, charitable trusts, and insurance contracts of all kinds. If one is investing for thirty to forty years into the future, nearby risk and returns may not matter a bit.
By 2005 it started to get through to me that I'd be retiring from work (for pay) in early 2008. Having lived through 1987-90 and 2000-2003, I began to develop some "healthy" fear thinking about what a really bad bear market just before or after retirement could do to me and my family. That's when and why I started this blog, and when and why I began cutting back even though I remained bullish on inflation beneficiary stocks and sectors.
Yesterday I exchanged emails with a younger friend who is a successful trader. I sent him the ten year chart above of those Vanguard funds, and he said that was highly selective, and if you looked over long periods things would look a lot different. I thought that was a good objection, and went to look at the last 20 years for the same charts. ( Investors FastTrackInvestors is my data base going back to 1988, and their software is what I use for indexes, mutual funds, and closed end funds.) I myself was shocked that the Vanguard Long Term Treasury Bond Fund still had the highest annualized yield over the past 20 years! Granted that the bond market has had the greatest bull market, after that bond bull market of the 1930's and 1940's, since 1981-84. But so did the stock markets have great bull markets until 2000.
In recent posts here I quoted and talked about Peter Bernstein's recent advisory opinion that perhaps stock yields could once again exceed bond yields as they did from the 1870's to 1958 but have not done so since 1958. Also I have presented my opinion that unless the nearby US T bond futures contract can get above and stay above 123--the recent September 2008 and the June 2003 T bond high--bonds were very likely going down for a long time to come. Also I have believed and written for some time that we are in a longer term inflationary era with a current "vacation from inflation". Again, only if the T Bond gets and stays above 123 would inflation possibly be dead and deflation become a real possibility.
In Hussman's 1999 post quoted from above, he said this about interest rates and stocks:
"From an analytical perspective, nothing affects the stock market except by affecting these three factors: dividend payouts, earnings growth, and P/E ratios. The primary effect of interest rates is on the P/E ratio. Falling interest rates encourage falling earnings yields (higher P/E ratios). Rising interest rates encourage rising earnings yields (lower P/E ratios). These effects are strongest when P/E ratios begin at extreme levels."
I added the large and bold font for the last sentence. If the last twenty years of falling interest rates saw the long term treasury bond annualized gains surpass stock gains, and if rates may well start to rise, especially if the on-going world-wide reflation attempts succeed, what will that do to stock returns for the next ten years? Stock P/E ratios have fallen this year, but they are still above the levels seen at many bear market lows. Higher interest rates would only increase the likelihood of lower P/E ratios, according to Hussman's last sentence, and that would mean a double whammy and lower stock prices. On the other hand, if interest rates fall substantially from here (TBond > 123, etc), that would imply deflation is coming and that the reflation attempts failed. That would scarcely be good news for P/E's or stocks either, but great for bonds.
If P/E's fall to 10 or below, as I imply above, annualized total returns for US stock indices could be only 4-6% even though stocks appear somewhat undervalued at this time. That would be even less than the annualized 8.6% returns for the S&P500 since 1988! But it would still be better than the returns or the past ten years of -1% per year.
William D. Gann's methods for capitalizing on extreme price movement have come into their own again in 2008. Gann of course published short term methods including the three day rule etc., but he made most of his money and most of his converts on the trading of extreme movements. If you've read Jesse Livermore and William Gann and know about market history from 1890 to 1932 you know why high volatility trading technique is important.
As we have all seen, most of the usual technical analysis approaches and valuation approaches are not timely or useful in very sharp sustained moves. Indicators and sentiment get "oversold" or "overbought" and stay that way. Normal inter-market relationships of different asset classes get torpedoed when everything is being dumped or bought at the same time.
One example of a Gann trade I discussed here in September had to do with the double top of the 2003 all-time high in the US Treasury Bond CBOT futures contract on September 16. http://tinyurl.com/6k7vd6 In all the comments on US T bonds I have seen since that date, only one other person very briefly and peripherally mentioned that crucial event, but you can bet that many experienced and powerful traders took note of it ahead of time as it approached and were there to sell. The bond market didn't pivot and turn there by accident above 123.
This past month we have seen a number of Gann hallmarks in the crash into October 10th. A very simple but powerful example was the date. Last year's high was October 11, and the 2002 low was October 10. Many Gann analysts call them "Ganniversaries" and never forget them.
In the past month I have heard from three Gann analysts, all three of whom I have known of and/or corresponded with for a decade or more, but not for quite a while. I take that alone as a "signal".
A very striking example of Gann methods was posted in October on an important internet site I visit by "Stoxx". If one draws a line from the 1932 Dow Jones low through the lower left hand corner to the upper right hand corner of a Gann box defined by the 2000 high time and 2002 low price, it hits the 2007 high price at October 2008. I am showing a quarterly chart for greater clarity, but the monthly chart is exactly the same. This is an example of what Gann always called time and price balancing.
Rick Lorusso of Citi-Futures sent me three SPX charts, the first showing the Fibonacci ratios in time of the 1974- 1987 lows and 1987-1988 lows and of the 1987 to 2000 high and 300 high to 2007 high. The second chart show the times to the October 2008 low from previous major lows and highs. The third chart shows the Fibonacci retraces of the 1994-2000 bull market and the range expansion targets off the 2000 and 2007 tops. (Gann did not explicitly use Fibonacci numbers, but his 1/8's of range apprimate the major Fibonacci ratios, and many of his timing intervals used Fibonacci numbers. Note that Lorusso's charts are shown here by courtesy of Citi-Futures.)\
Another Gann analyst,TAF, told me about the Gann angle from the zero price line at the time of the 2003 secondary low. Gann always drew various angles from all-time lows or highs, but he also drew angle lines from zero price or baseline at important secondary lows and highs. TAF has a better understanding of Gann angles and their proportions than anyone I have known. He told me to draw a line on a monthly chart of SPX up from the zero price point in March 2003 at 12.5 points per month. He has this on his own hand drawn chart for a long time. 12.5 point times 67 months came to 837.5 in October 2008. The October 10 low was 839.80.
These are methods in the folklore of trading for generations. They preceded Gann but he learned about them and publicized them. They are still largely unknown but are powerfully used by large and powerful traders. All of these studies suggest that time and price are right for a substantial rally in stocks.
Peter Bernstein, one of the most senior investment advisers currently in practice, and the author of many excellent investment books, has recently presented an argument that the long-standing relationship of stock yields and bond yields may be close to reversing. Before 1958, stock dividends yields had been higher than bond yields since the 1870's. Bernstein recalls that his senior partners in 1958 were convinced this was a short-lived anomaly, but that it has persisted to the present. After all, stocks were inherently more risky than fixed income bonds, in the then current and long time view, and so they should pay a higher dividend. But during the sharp recession of 1958 (-10% rate of change in GDP), inflation increased and bond rates finally rose above stock dividend rates. It was unheard of for inflation and bond yields to increase during a sharp recession.
Bernstein understood that the growth stock ideology was just getting started, indeed he had trumpeted it two years earlier in the Harvard Business Review. The new ideology, which we now accept as eternal truth, was that growth stocks would have superior and accelerating cash flows and dividends far beyond a fixed rate bond so that current stock dividends were immaterial, so they should pay a smaller dividend than bonds. And so it was. Until 1999. Since 1999 the ten year US Treasury yield has fallen from about 6.5% to 4% while the SPX yield has risen from nearly 1% to nearly 3%, and recent SPX price losses may boost that yield to 4%.
However, the second part of Bernstein's observation is that during the inflation that followed the 1960's, SPX dividend growth kept up with inflation as did bond yields. Mind you that as a result of this stocks rose until 1972-73 generally, while bonds fell. The positive correlation of bond and stock yields continued all through the 1980's and 90's as both yields fell and both stocks and bonds therefore rose. But as we have seen above, this trend has reversed since 1999-2000 when inflation rekindled.
Bernstein now wonders if current difficulties in the economy will result in stocks falling far more and Treasury bonds rising far further (yields lower), in which case stock dividend yields could far exceed Treasury yields for the first time since 1958. Furthermore, he points to the lag of inflation-adjusted US incomes compared to increased productivity as a major political issue which may impede corporate earnings and lead to much lower stock prices. He implies this would be the reason for a reversal of the 1958 crossover of Treasury yields above SPX dividend yields.
Bernstein's analysis is very subtle and extremely perceptive as only a seasoned observer can be, far beyond most current commentators. Personally, however, I believe the feasible rebuttal lies in the same inflationary experience that began in the late 1950's and which has begun again since the turn of this century, namely inflation itself. Stocks may not do as well during inflation as during benign disinflation, but they do better than bonds. So earnings, after the recession, and dividends, will rise, but probably not as fast as Treasury bond yields, so stocks will be a better investment than fixed income Treasury bonds, just as they were in the 1960's and 1970's. This is totally compatible with the Economic Long Wave which predicts the same outcome. Inflation will decimate bonds but earnings and dividend growth of substantial corporations will partially protect them. For better protection one will want to be invested in inflation hedges such as selected currencies, metals, and commodities either directly or via equity-linked vehicles.
The first chart is the computed seasonal or annual cycle of the Dow Jones based on daily data from 1928 to 2008. The algorithm is proprietary to CSI Data, my data service for decades. Other providers show similar charts.
Cyberfriend and super market analyst Christopher Carolan has another method of analyzing the annual cycle, which is also proprietary but which has a good record for periods when I have seen it.