Doing What It Needs To

Bottom Line: The week after tax day has been quite positive the past 7 years, but historically there’s no reason to expect that to continue.

I often get requests to check if various statements from the media, or common market lore, are correct. One of the statements that has been making the rounds lately is that the week after tax day is usually positive. Like most things that garner attention, this is true (but only a little bit).

For the past 7 years (and 8 out of the past 9), the Dow Jones Industrial Average has been higher 5 days after the tax filing deadline every time, with an impressive average return of 2.4%. I think that’s where the impression has been formed about the week after tax day. Unfortunately, that’s also about where it ends - by going back further, it becomes readily apparent that the week is no more or no less positive than any other week.

Using the DJIA, we can go back to the beginning of when personal income taxes were placed in effect. The filing deadline was March 1^st beginning in 1914, changed to March 15^th in 1918 and finally changed to April 15^th in 1955. Looking at the entire history of the week after tax day, the DJIA was higher 57% of the time, with an average return of 0.3% - almost exactly matching any average week during the past 90 years.

The table below outlines how the DJIA performed 5 days after tax day, from its inception through last year. While the performance given in the table shows 5 days later, the results are consistent for every time frame studied. If the filing deadline fell on a weekend, the next trading day was used as the starting point.

The Week After Tax Day DJIA, 1914 - 2003
Decade	5-Day Return	% Positive
1910's	1.6%	67%
1920's	0.1%	60%
1930's	-1.1%	40%
1940's	0.4%	50%
1950's	0.6%	80%
1960's	-0.6%	40%
1970's	-0.1%	40%
1980's	1.0%	80%
1990's	0.4%	40%
2000's	2.6%	100%

I cannot find any consistent pattern among the decades. The week following tax day wasn’t any more or less positive 60 or 70 years ago than it has been recently. While the latest streak of the past 7 years being positive is unusual, it is not unprecedented. We saw 9 straight years in the late 1970’s through the mid-1980’s when the week was positive. There were a couple of other streaks of 5 or 6 consecutive years showing positive weeks as well.

Oftentimes, I think there is as much value in finding out if something is not true than finding out if something is true. Both can help improve results, or at least prevent unnecessary losses. I think banking on a positive market over the next week simply because that’s what has been reported is dangerous. While the current streak could continue, it is important to realize that there is no historical precedent suggesting the coming week should be markedly different from any other.

Bottom Line: A look at sentiment in the bond market suggests we may need more pessimism before a good bottom, but if the recent relationship between the two markets continues, such a bottom would not necessarily be good for stocks.

With all the focus on the movement in the bond market, I thought I would take a few minutes to show a snapshot of our sentiment measures which track the 30-year Treasury Bond. Since 1977, the correlation between the S&P 500 and 30-year Bond yields is -0.76 (on a scale of -1.0 to +1.0), an extremely strong number considering the length of time studied. A negative correlation means that as one rises, the other falls. Since we’re talking about yields, this means that as the stock market has risen, yields have fallen, and bond prices have risen along with stocks, i.e. the stock and bond markets have generally moved in the same direction since 1977. However, beginning in January 1998, the relationship changed and the two markets began moving in almost exact opposite directions – the correlation between the S&P 500 and Bond yields over the past six years has been +0.72. Again, this is extremely strong given the large number of data points.

Obviously, the relationship has not been perfect, and there are times when the two markets simply march to their own tune. Many seem to be hailing the surprisingly positive jobs number on April 2^nd as a seminal event in the Bond market, signaling the (much anticipated) bear market that so many experts have been expecting. If we look at how stocks and yields have moved since that report, the correlation reverts back to its old ways at -0.80, suggesting that for now, what is good for bonds is good for stocks. Apparently, equity traders (for the moment, anyway) are concerned about higher interest rates, so if we get bad economic numbers or what have you, and bond prices rise, stocks may as well.

There’s really no telling if the most recent relationship between stocks and bonds will hold. The long-term positive correlation between the two markets is extremely strong, but the negative correlation between them since 1998 has been just as strong. If the experts on the Bond market are correct, and yields are about to head significantly higher over the coming years, then stock bulls had better hope that the intermediate-term relationship between the two markets keeps its negative correlation.

We know that there is a strong link between the stock and bond markets, so it may help stock traders to see how bond traders are perceiving their market. First, let’s look at where traders are placing their bets in Bond futures. On the site, you can find the separate positions for both commercial hedgers (i.e. large traders) and small speculators. For purposes of brevity, below I have included only the “variance”, which is simply a combination of both sets of positions. When the variance is high, that means commercial traders are net long and/or small speculators are net short – a configuration which normally results in rising bond prices and falling yields. On the other hand, a low variance would indicate that commercial traders are quite net short and/or small speculators are net long – a bearish indication for bond prices.

The blue arrow on the right scale shows the current reading of the variance. At +43,000 contracts, the indicator just entered “extreme” territory by being more than two standard deviations above its long-term mean value (shown by the thick green horizontal line on the chart). This is not nearly any kind of record, but it is unusually high, and it is due to both commercial traders being quite net long and small specs being very net short. Also notice from this long-term chart that the uptrend in bonds is still very much intact (for now).

Next, let’s look at investor opinions towards the bond market. Both Market Vane and Consensus, Inc. poll traders on the bond market, and report their results weekly (though Consensus does not release theirs publicly until an extra week has passed). The chart below shows where these opinions are in relation to some past readings.

As of the most recently released public data, both surveys show trader opinions about in the middle of their long-term ranges. Consensus, especially, has shown a dramatic drop-off in bullish opinions over the past couple of weeks, but neither poll is yet showing what I would consider to be excessive pessimism.

Now let’s switch to another measure that shows where traders are putting their money – the options market. Options traders on bonds tend to be better market timers than their equity market cousins, so put/call ratios here are not quite as effective a contrary indicator. Still, overall, I think it helps to view the data on a 10-day or 21-day moving average basis. The chart below shows the 21-day moving average of the put/call ratio on 30-year Treasury Bond futures.

Once again, this data is about in the middle of its range. While there have been at least 1 ½ puts traded for every call on 4 out of the past 5 days, pushing the 10-day moving average into extreme territory, not enough time has passed to move this longer-term average into oversold territory. Excessive put volume over a month’s time has been a pretty decent indicator of a bottoming Bond market over the past few years, so watching for this indicator to move into extreme territory would be welcome news for bond bulls.

Lastly, we’ll look at another real-money measure – the flow of assets into the bond funds at the Rydex mutual fund family. Rydex carries two funds concentrating on Treasury Bond movements, one that profits when Bonds rise in price (the Bond fund) and another that profits when Bonds fall (the Juno fund). As of Friday, there was $50 million invested in the Bond fund, compared to an astounding $2.02 billion in the Juno fund. Never before have assets in the short-side fund been 40 times what was in the long-side fund, as we’re seeing now. The chart below shows the “flow” into the two funds, with green bars highlighting those times when bullish assets dropped to a very low level and/or bearish assets became exceedingly high.

If you look at how bonds performed after the green bars, it was quite positive, as bonds normally rallied soon afterward. Currently, bullish flow is reaching an oversold extreme, nearly on a par with other lows over the past few years. With such a large asset base, however, it is becoming more difficult for the bear flow to match that type of pessimistic extreme. While the bear flow is becoming high, it is not yet in what I would consider extreme territory. It may seem ridiculous to suggest that over $2 billion in bearish bets placed with a somewhat obscure mutual fund family is not a display of excessive pessimism, and from a common sense perspective it’s hard to argue with that.

As a recap, the most recent data available is showing commercial traders net long Bond futures (though not to a truly excessive degree), small speculators net short, put volume becoming heavy over the past few days (but not yet enough to push some of the moving averages into extreme territory), the sentiment surveys showing a quick drop in optimism but not enough to give any indications of an oversold Bond market, and Rydex traders moving a significant amount of money to the short Bond fund. Taken all together, it does seem as though sentiment on bonds would support something of a rally at any time, but we’re not seeing the type of all-out panic seen at most of the other lows over the past few years. Also, if the long-term trend in Bonds has changed to down as so many experts are predicting, then we haven’t seen anything yet. If that is the case, then I would expect to see much, much more pessimism in the indicators posted above, and even then Bonds should show only minor bounces.

If you believe that sentiment in the bond market has reached a pessimistic extreme, and if you believe that the recent re-coupling of those two markets over the past two weeks is going to continue (a couple of big “ifs”), then that would suggest you should look for a rally in equities as well. Personally, I think that while there is some evidence that pessimism on bonds has reached a peak, I don’t see the types of readings that have lead to major bond lows in the past – and that’s assuming that the long-term trend in bonds is still up. As for the correlation between the two markets, that’s a much more difficult question. While I always try to respect historical relationships, I tend to give more weight to what has happened recently (like over the past few years). So I would expect stocks and bonds to once again move in opposite directions, as they generally have over the past six years, but I am not exactly comfortable with that assumption and I will be watching their interplay very closely.

On Wednesday, I noted that I would be watching the 1120ish level on the S&P as a place that should see a reversal higher if we were going to. We hit 1120.75 on Thursday before the S&P turned and did what it was supposed to, so I have to continue to believe that the uptrend is still intact. My suggestion in that comment was that the next few days could be a good tell for the longer-term outlook, and as far as I’m concerned the market has passed the test so far. An immediate failure from here would not be a good thing of course, but at the moment I don’t have any decent reason to expect one. Nearly all of our short-term measures are back to neutral, though the 3-day cumulative TICK I mentioned on Wednesday is still working off the near-record oversold extreme it hit late last week. I’m still watching 1120ish on the S&P as the “canary in the coal mine”, and so far it’s looking healthy, so my preference continues to be trading from the long side.

This disclosure is not intended as trading advice in any form. It is meant as a note to subscribers that the author may have a position directly affected by the market outlook reflected in the commentary. Although the author takes great pains to remain objective in any commentaries, it is only fair that readers should know that the author may have taken positions in accordance with his market outlook. Positions can and do change at any time, without notice to the reader.