Anyone who thinks forecasting market direction is a relatively easy process in today's manipulated, policy driven environment is ill informed. Certainly history provides no period we can reference that resembles where we are today. The reason is simple - we have never in history had such an active period of fiscal and monetary intervention that has created such distortions in the various asset classes.
That said it is still worth the effort to attempt to reason out where we might be headed in the coming weeks and months and the logical starting point is to look at where we are today. The following chart is the YTD comparison of stocks, bonds and precious metals using SPY, TLT and GLD as proxies for the 3 asset classes:
Stocks are up roughly 20%, gold down 20% and bonds down 10% YTD. Here is what I see as the most logical reasons for each asset class being where it is today:
- Trailing earnings support the current price.
- Fed policy has been highly supportive of stocks.
- Money flows moving out of other assets have pushed PE multiples.
- Inflation expectations have not materialized.
- Low interest rates have pushed money out of bonds and into equities.
- Talk of Fed bond support being withdrawn.
To put things in perspective it is useful to take a look at a little longer view of these asset classes. The following chart is the 5 year version of the chart above:
From a 5 year view all 3 asset classes are still in positive territory. Stocks are up roughly 30%, gold is up about 60% and bonds are up just short of 20%. From a forecasting perspective it is a virtual certainty that we will at some point revert to mean as we always do. The question though is where will mean be at that point. Here is a longer term view of SPY with the mean value calculated as the 500 period moving average:
The mean value is 141.09 but as one can see from the chart above as often as not when the momentum shifts are sufficient to move price to the mean it usually moves on down to the -2 standard deviation Bollinger band. That level is 112.28. The long term chart is useful but not so much for short term forecasting. For short term forecasting the 50 day chart is more predictive. Here is a look at the 50 day chart with Bollinger bands included:
The chart dates back 5 years. I have drawn a line from each peak to each trough as it moved back and forth between the +2 standard deviation Bollinger band and the -2 standard deviation band. There are a total of 20 trips from peak to trough - 10 round trips - in the 5 years reflected above. That's an average of 2 round trips per year. In fact a close look at the chart shows that in each of the years reflected above the market moved back to the lower band exactly 2 times.
One of the most useful tools I know of for predicting when the market will shift back to the lower band is based on momentum. In other words when momentum is interrupted to a degree that is sufficient to move the market a full standard deviation in the direction opposite the trend it generally signals a temporary momentum shift that is significant enough to push back to the lower end of the range.
The 3 month chart below demonstrates this phenomenon:
The chart above doesn't plot the 2 standard deviation band - rather it plots the 1 standard deviation band. It is the move from the outer band through the inner band that is useful for predicting momentum shift. The momentum shift points are reflected above. Following these signals one would have sold on May 31 @ 164.19, bought on June 25 @ 158.57 and sold again on July 26 @ 167.70.
Two points to make here. The first point is that on occasion the method used gives false signals and therefore a 1 standard deviation stop is suggested. The second point is that the longer term trend cannot be identified using the short range chart. If we stretch out to the 200 day MA and standard deviation calculations we get more information on the intermediate trend. Here is the chart:
What's most useful here is that to date we haven't had enough upward momentum to push back through the +2 standard deviation band on the intermediate term chart. In other words we now have both the short term signal and an intermediate term signal to the short side of the trade that is still in play. Here is a look at the chart above with the 1 standard deviation bands plotted showing that momentum was sufficient to give a sell signal at the June lows and the rally off those lows wasn't sufficient to move back to the +2 standard deviation bank stopping the trade out:
Here is the longer version of the chart showing the 500 day numbers:
This chart plots the 1 standard deviation bands and indicates that from a longer view we are still long the market. In other words we never made it back to the +1 standard deviation band and therefore long term traders using this chart would still be long.
Let's summarize where we are then. Based on the short term 50 day chart we are short the market at 167.70 with a stop at 172.75 and a target objective of 161.60. Based on the intermediate term 200 day chart we are short the market at 159.08 with a stop at 173.72 and an objective of 135.10. And on the longer term chart we are still long the market and will remain so until the +1 standard deviation band is penetrated at 155.40. If we get a sell signal on the 500 day chart the target objective will be 112.47 which represents a 34% pull back from current price.
For a real long term view check out this chart of the inflation adjusted real S&P 500:
Reversion to mean looking at the longer view would put the S&P 500 at an inflation adjusted 422. Before completely discounting this as nonsense here is Doug Short's long term version of the S&P 500:
Here is a comment from Doug relating to this chart:
Historically, regression to trend often means overshooting to the other side. The latest monthly average of daily closes is 67% above trend after having fallen only 11% below trend in March of 2009. Previous bottoms were considerably further below trend.
The 11% below trend was a low on the S&P of 667 and fell well short of historical regression to trend levels. In other words the regression to mean - when it occurs - may be much below what almost anyone sees as plausible at this point. But then the fall from the 2007 high to 667 was also not seen as plausible in 2007.
Let's move on to the bond market. Here is a look at the long view:
From a statistical probability perspective this market has moved down about as far as we might expect in the short run. Here is a 50 year look at the 10 year yield to put things in perspective:
We are still at very extreme lows from a historical perspective but from a statistical probability perspective we are at that point where the odds of moving much lower are less than 5%. In other words in the near and intermediate term it would appear we have done about all the damage to the bond market we are going to do.
Let's take a look at the short term 50 day chart to see where we are from that perspective:
The chart above is the 50 day chart with the 2 standard deviation bands drawn. The next chart is the same as above with the 1 standard deviation bands:
The chart above shows that a long trade would have been entered at 109.03 using this methodology with a stop at 103.93 and a target price of 115.31. The next chart is the intermediate term 200 day chart:
The intermediate term 200 day chart indicates a market that is moving sideways as the MA slopes lower pulling the channels downward. In other words both the long term and the intermediate term chart show the market is still in a downtrend and trading off these charts one would remain short bonds. The 50 day chart has given a buy signal with a maximum upside objective of 115.31.
If the 115.31 level is reached the intermediate term 200 day chart will also generate a buy signal. Here is the 200 day showing the 1 standard deviation bands:
The take away from all this on the bond chart is that we are probably going higher in the very short term. We've moved to the outer lower bands on the long term, intermediate term and short term chart and we have been given a buy signal on the short term chart suggesting that we may get a little bounce to the upside but we still remain in a cyclical bear market. The odds of going much lower in the near term aren't very probable though as we need to allow the channels to shift lower. We are trading at the -2 standard deviation bands and the probability of moving significantly lower is 95% against that occurring.
On the other hand from a very long term perspective we are at historically low yields but that is as much a function of monetary policy as anything else and the odds of the Fed shifting their stance in this area are very remote at this point. Furthermore we have some fundamental conditions that are working against the bond market at the present. First, the talk of tapering is not just talk - they will do it.
A good part of the reason for taper is the on-the-run Treasury issues are being reduced in quantity as the deficit is being reduced through sequester spending cuts and projected revenues from tax increases. For the Fed to continue at the current rate of purchase on Treasuries would mean they would be taking the equivalent of approximately 70% of new issues.
The second consideration for the bond market is that foreign sales of U.S. Treasuries had been surging in recent months but that situation has reversed and Treasuries are once again drawing foreign purchase interest. This from Bloomberg on the matter:
Treasury 30-year bonds rose for the first time in three weeks as U.S. auctions of $72 billion of securities including long bonds at two-year-high yields attracted above-average demand from foreign investors.
"The week's auctions were very strong and there has been more reluctance for investors to extend into riskier markets, which also supports Treasuries," said Christopher Sullivan, who oversees $2.1 billion as chief investment officer at United Nations Federal Credit Union in New York. "The yield rise has been arrested to some degree, as the possibility of Fed tapering is now largely priced in."
A third consideration is that bonds at current yields are really pretty attractive relatively speaking. Here is a look at the dividend yield of the S&P 500:
Compare the 1.97% dividend yield of the S&P 500 with the 10 year Treasury at 2.58%. Also look at where we are on the charts above on the S&P 500 relative to the bond market. The bond market has fallen dramatically and at least in the short run isn't likely to move much lower. On the other hand stocks are now at all time record highs and suggest a market that is overextended. The risk to principal loss in stocks is much greater than the risk of principal loss in bonds at these levels.
A 4th consideration is the safe haven aspects of the Treasuries of a reserve currency nation. Although taper talk has caused bonds to fall dramatically in recent months the odds of further significant declines at this point are statistically improbable and the odds of stocks falling from these levels are very high from a statistical probability perspective. If stocks do fall significantly bonds will receive the benefit as money is moved out of equities and back into bonds.
Bottom line - don't look for bonds to fall much further in the coming months and if stocks do sell off significantly the odds of bonds moving back into a cyclical bull trend are very high. We do have a short term buy signal in bonds and a move back above 112.00 on TLT would give us an intermediate term buy signal. My guess is that Treasuries are a pretty good bet to move higher in the coming weeks and months.
Now on to gold GLD - here is the long term chart reflecting the 500 period Bollinger bands:
The long term chart shows that gold is significantly oversold. Let's see what the intermediate term chart looks like:
The 200 day chart shows that gold is basing - at least temporarily. A look at the same chart with the 1 standard deviation bands plotted will tell us if we have had enough of a counter trend move to give us a buy signal:
The intermediate term chart shows that gold still hasn't moved enough to give us a buy signal. The close at 126.86 on GLD is still below the 132.44 -1 standard deviation band and therefore we would still be short gold based on the intermediate chart.
Here is a look at the short term chart:
On the short term chart we do have a buy signal at the -1 standard deviation band. Here is a look at the same chart with the 1 standard deviation bands plotted:
We did get a buy signal on the short term chart on June 11 @ 123.90. Additionally, and perhaps significant as far as the longer term perspective is the island reversal at the bottom of the chart.
The analysis above makes sense in every respect to me. I do think we are way overdo for a correction in stocks but as I noted above there are reasons that support stock prices at current levels - in particular from a backward looking perspective.
Additionally, although not entirely rational in my opinion - the bond sell-off can at least be explained from a fundamental view. First of course is the taper talk. Second is the move to by-pass the U.S. dollar as a reserve currency with a number of bi-lateral trade agreements amongst the BRIC countries. This has prompted selling of reserve assets by foreign countries and put downward pressure on bonds.
Going forward the shift in trend in both the bond and stock market makes sense as well. We know that stocks are at pretty high PE levels and we know that profit growth is contracting. We also know that we are in a period of disinflation and very close to deflation and we know that GDP is contracting. We now have a 3 quarter average of less than 1% on GDP. None of that bodes well for stocks going forward.
A trend shift in bonds seems likely if for no other reason than a stock market correction which seems imminent at this point. The yield at current price on bonds - when compared to stocks - seems certain to drive assets into bonds going forward. Additionally, even though U.S. Treasuries have been sold off in recent months by foreign countries the appeal of Treasuries as a safe haven asset in a stock market sell-off is driving foreign investors to move back into bonds.
All of that makes sense but the gold market is also suggesting a shift in the cyclical trend and that is not as easily explained. Gold is a function of inflation and deflation is the more likely short term scenario. Here is a look at the CPI:
And here is a look at gold over the same time frame:
For those who don't think the rate of inflation has been the driver of lower gold prices the two charts above suggest they are wrong. The truth is inflation using the CPI as a measure peaked in mid 2011 and so did gold. They have both moved lower since and it is not logical to assume that correlation - although not perfect - hasn't been the driver of gold.
The point is simple - unless the private sector banks suddenly reverse course and begin to lend at a much increased pace the current trend on inflation is likely to continue and deflation is a very real possibility. In fact if a recession enters the picture in the coming months the odds of banks ramping up lending are remote and the odds of recession are reasonably high. The truth is we do have them and fairly often. We've had 11 recessions since 1950 - one about every 5.5 years on average.
Still, it seems to me that we are probably at a point where gold is building a base and likely to move higher in the coming months. If my analysis is correct it will do so even though we continue to see disinflation and possibly deflation. Is there an explanation that would allow higher gold prices that diverge from inflation? Perhaps so.
Gold seems to be at the forefront of a global controversy with what appears to be a high demand for the physical commodity as the price struggles to find support. There seems to be little doubt that the BRIC's are accumulating gold. It is also a fact that the BRIC's are working together to circumvent the U.S. dollar as a reserve currency through bi-lateral trade agreements with one another.
Here is a short list of articles on the subject for those who are interested:
- Australia to Bypass US Dollar in Trade with China
- Emerging powers China, Brazil make plans to bypass U.S. dollar
- Yuan Replaces the Dollar in China's Dealings With France, Britain, Australia, as the War-Debt Continues to Destroy US Currency
- China, Australia, Japan, Russia, Iran, India and Brazil have ditched the US dollar: are we witnessing the dollar's last days as world reserve currency?
Here are a few articles dealing with the increase in gold demand amongst the BRIC's:
- Brics Are Buying Gold Case Study:India
- New Gold Standard to Emerge?
- Countries Buying the World s Gold
This is happening and to suggest that the global landscape in the next few years with regard to the U.S. Dollar's reserve status will be the same as it is today is to deny the facts. In other words it is possible that gold will no longer be a function of inflation and the devaluing of the US dollar so much as it is global demand for gold as a reserve asset.
So where are we headed from here? As is usually the case I find myself in opposition to the majority. I don't think we are in an economic recovery that will provide further fuel for higher stock prices. To the contrary I think we are on the verge of a long overdue correction that may end up being much more severe than most think.
And I don't think interest rates are going to normalize. What I think is that U.S. Treasuries are at the bottom of a cyclical bear market in a longer term bull market and that we will see rates move lower and bonds move higher from these levels. I think the flight to safety may end up driving U.S. Treasuries to new highs in the coming months.
I do see a deflationary scenario developing which should cause the dollar to hold value - if not increase in value relative to other currencies but in spite of that I see gold moving higher from these levels. In other words I think gold will move higher but not because of a weak dollar. Rather I think gold moves higher based on a high demand for the metal as a pseudo reserve asset.
That is not the consensus view amongst the gold bugs who expect gold to soar due to rampant inflation and a devalued dollar borne from the excesses of QE. That may well come in time but it is a long way off as we must surely go through a period of contraction first.
Bank lending won't occur until bankers and borrowers perceive assets as being undervalued and that is not the case today. These aren't easy times to forecast markets. We are in a period unlike any other in our history. It would appear - at least to me - that we are about to take the next step toward a global economy that seems certain to transform the playing field in a way that most don't expect or understand.
What I see as the ultimate outcome is a non-sovereign reserve currency that allows nations with high debt levels to monetize those debts by a process of devaluing their respective currencies and that must necessarily happen through the private sector as a process of expanding money supply through the fractional bank multiplier. That is not likely to happen as long as those nations currencies are used as reserve currencies and that includes - to some degree - the eurozone and Japan.
In the end I put great stock in the predictive value of statistical probability - especially when the market moves to the outer Bollinger bands and then reverses trend by a full standard deviation. That is what we see today in all 3 markets and the short term charts have now moved in the opposite direction of the trend by a magnitude sufficient to create a buy signal in bonds and gold and a sell signal in equities. My advice would be to follow those signals.
Disclosure: I am long UVXY, FAZ, TZA. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.