Let the Fed-Guessing Begin
Market participants weren't quite sure what to expect as this week began. Many indices ended the previous week with key reversal days, but the bulls weren't through yet. Although key reversal days come after a seemingly never-ending rise or plunge, and so are closely watched, they're often minor reversal signals. That's what last Friday's key reversal day turned out to be: a minor reversal signal.
With most indices, the reversals stopped at the midline support of their ascending regression channels and the indices continued to find support there through the week. Neither overbought conditions nor Thursday's worrisome four-week initial claims average could drive the indices back below midline support. In fact, up until Friday, trading felt like the old momentum days of late 1999 and early 2000, complete with news of mergers and acquisitions, and IPO's. FormFactor (Nasdaq:FORM) began trading Thursday. With MSDW as the underwriter, and Lehman, Bank of America, and Thomas Weisel as secondaries, 6 million shares were priced at $14.00, up from the 5.5 million originally priced at $11-13. FORM designs, develops, and manufactures tools used in semiconductor manufacturing, specifically wafer probe cards that allow advanced tests and speed the manufacturing process. The issue opened Thursday at $19, but ended the week at 16.50, still above its $14.00 pricing. FORM's status as a semiconductor-related stock ties it to the performance of the semi industry and that industry's link to global economic recovery. This week, the price of the 256 Mb 400 MHz DDR (double data rate) chip continued recent increases, with the price surging 5.5 percent on Thursday alone. The Semiconductor Industry Association pegged the 2003 global chip sales growth rate at 10.1 percent, while forecasting 16.8 percent growth in 2004, 5.8 percent in 2005, and 7 percent in 2006.
The semi news wasn't all good, however, perhaps producing the SOX's almost 30-point fall this week and an evening-star formation on the SOX's weekly, rather than daily, chart. While 2003 global sales are expected to grow, the SIA forecast 2003 Americas sales to decline 2.1 percent, with growth resuming in 2004 at a 15.7 percent rate. Deutsche Securities added further pressure to the semis with its downgrade of Intel (INTC) on valuation concerns. INTC fell 3.53 percent in Friday's trading.
In addition, news surfaced late in the week that Taiwan Semiconductor (TSM) and STMicroelectronics (STM) were being added to the SOX, with Lattice Semiconductor (LSCC) being dropped. This marked the first time the index would include U.S. shares of foreign companies. Some market pundits at least partially attributed the SOX's weak performance this week to confusion as to how their inclusion would affect the index. Of the mergers or acquisitions announced globally, Oracle's (ORCL) hostile bid for PeopleSoft (PSFT) probably drew the most press. Since announcing its cash takeover bid of PSFT the previous week, ORCL fielded questions about its intentions toward PSFT. Analysts speculate that SAP might benefit if ORCL's bid succeeds and ORCL attempts to transition PSFT's customers to the ORCL products. This week, PSFT rejected the hostile bid and announced intentions to continue with its acquisition of J.D. Edwards (JDEC). ORCL moved up its earnings release, with many rightly speculating that ORCL would surprise to the upside. ORCL closed the week at 13.48, up 0.62 this week. Friday, the buoyant 1999-ish mood dipped as attention focused on a full slate of economic releases. The before-the-bell releases included May's PPI and trade gap, and the 9:45 release featured June's preliminary consumer sentiment number. The consensus forecast for the PPI number was a decline of 0.1 percent, with estimates as low as a decline of 0.3 percent. The actual number fell to the lowest of the estimates, declining 0.3 percent.
Economists predicted a 0.1 percent rise in the core number, with estimates ranging all the way to a 0.3 percent drop, with the actual number meeting the expected 0.1 percent rise. April had seen a decline of 1.9 percent in the PPI number and a decline of 0.9 percent in the core number. Analysts surmised that Friday's PPI number left all options available to the Fed, as the low inflation risks allow a further easing in the rates.
While the dollar's plunge should help narrow the U.S. trade gap, making our goods more attractive overseas, that effect was minimal, at least in this reporting cycle. The April trade gap did narrow to $42 billion, as expected, down slightly from March's $43.5 billion, but still remains up 26 percent from last year's rate. Exports fell 2.2 percent, but an April decline in crude oil prices also sent imports down 2.1 percent. Economists predicted a trade gap of $42 billion, but estimates ranged as high as $46 billion, with March's gap at $43.5 billion.
Although this number isn't rated a market-moving number, economists do hope to see rising exports, a hope that was dashed in this reporting cycle. Although the U.S. does not depend greatly on exports, rising exports can be seen as a sign of growing demand in other countries and a sign that their economies might be strengthening.
The University of Michigan's U.S. consumer sentiment number attracted the most interest, however, both in the U.S. and globally, both before and after its release. The June number disappointed, falling to 87.2. The consumer sentiment number had been predicted to rise to 93.2 from May's 92.1, although forecasts ranged from 91 to 93.7. None of the forecasts ranged into the 80's, however.
As Jeff Bailey mentioned in an intraday update Friday morning, consumer sentiment usually improves when stock markets gain, as ours have been doing. However, consumers mentioned worries about job losses. The component measuring future outlook also plummeted, to an unexpected 84.2 from the previous 91.4.
The low employment figures and consumer sentiment number increase speculation on the size of the rate cut that might be expected at the June FOMC meeting. Over the last weeks, Fed watchers have transitioned from asking whether there will be a cut to whether it will be 25 basis points or 50. The May retail sales, released Thursday, helped boost the case for a Fed cut. May retail sales met expectations, but the retail sales figures excluding autos were up only 0.1 percent, less than the projected 0.2 percent increase. That turned out to be the best of both worlds, however, as the small increase still allows the Fed to lower rates again while being just large enough to support hopes of a recovery in progress. Thursday's initial claims number demonstrated a continuing softness in employment, also buoying hopes for a rate cut.
Don't forget that economists across the globe will be debating the Fed's likely decision next week, and their markets' reactions can impact ours. Global bourses have been building on the gains of the others as hopes have risen that economic recovery is either already underway or about to be seen. For example, the Nikkei is up 15 percent since hitting new 20-year lows day after day in April, and it's gained 7 percent just this week. The FTSE 100, CAC 40, and DAX have bounded above resistance level after resistance level, but they've based that energy on hopes of a U.S. recovery to a great degree. To demonstrate the interdependence of the global bourses, European markets began falling as our economic numbers were released Friday morning and then followed the course of our markets to a great degree. The performance of European and Asian markets Monday morning may govern the direction of our opening, too, although after U.S. trading begins, it's often our markets that guide the directions of the others, rather than the other way around.
While Friday's decline felt strong and created potential reversal signals on the daily charts, it didn't do much technical damage. Many, me included, await the natural and expected retracement or consolidation of strong gains, if only for new bullish entries, but we've waited in vain so far. Many felt the retracement had begun in mid-May when the DJI fell more than 300 points in three days, but even as the DJI was plummeting, signs of renewed strength began appearing on the daily oscillators. Those signs included an upturn in the RSI, often a leading indicator, and then an RSI breakout above its trendline of descending highs. The daily chart depicts a bullish cross of the +DI and -DI lines on the ADX, and then an upturn above 20 on the ADX itself. May market behavior is old news by now, except that many of those chart characteristics remain intact as depicted in the chart below.
Daily Chart of the DJX, as a proxy for the DJI:
A study of the chart reveals that a potential reversal signal, an evening-star pattern, formed over the last three days' trading. It may portend a drop, but during this rally, the drops produced by the reversal signals themselves have often been the only declines seen. Today the DJI bounced from above the midline support, hinting that the support could hold even if the DJI were to decline again on Monday.
The DJI could fall as far as the 8880 level, another 300-point decline, and still remain within the ascending channel and above the rising 21-dma and the 8880 horizontal support.
A drop below 8800 would begin to damage investor sentiment and might predict a deeper drop, perhaps at least the 25-38.2 percent retracement we've all expected. The shallower 25 percent retracement would bring the DJI back to the next support level at 8720-8750.
Currently, the ADX remains above 30, however, and it's rising, showing that the trend increases in strength for now. For those not accustomed to studying the ADX, I believe Jane Fox is writing an article about the ADX for this weekend's newsletter.
The NDX daily chart demonstrates some of the same traits, although subtle differences appear. As the NDX printed its own version of an evening-star pattern, it did not bounce from the midline support of its ascending channel, but rather closed right on that support. RSI turns down more distinctly. The 21-dma (pink line) rises higher beneath the NDX, so that a move to the bottom of the ascending channel would violate that rising 21-dma. ADX still depicts a strengthening trend, as it did on the DJI.
Daily Chart of the NDX:
A fall as deep as 1140-1160 would maintain the integrity of that ascending channel, although it might damage investor sentiment more since a fall that deep would also carry the NDX below its 21-dma. A study of the chart shows that this index has more routinely violated that dma during the rally than has the DJI, however.
The downturn in the RSI, completing yet another instance of bearish divergence with prices, perhaps predicted today's downturn, but watch for a penetration of the rising RSI trendline for a first sign of a possible continued decline. Those bearish divergences will eventually mean what they used to mean, but for now they tend to predict a pullback only as far as the bottom support of the regression channel.
The OEX chart shows the same characteristics, with signs of renewed bullishness in mid-May as the OEX fell to the bottom of the regression channel and began moving up again, and with an ADX level that currently remains above 30.
Daily Chart of the OEX:
While not as classic as that of the DJI, the OEX chart reveals a close approach to an evening-star pattern, but it also shows the OEX springing up from that midline support, as did the DJI. Although it's not included here for space reasons, the OEX's 60-minute chart shows oscillators turning up out of levels indicating oversold conditions, too, hinting that the OEX could rise Monday morning if the performance of global bourses does not damage investor sentiment too strongly.
If those hourly oscillators and that spring from midline support send the OEX up on Monday, OEX 505 and then 507-508 might be appropriate places to watch for the next pullback toward the midline support of the OEX's regression channel. OEX 507-508 marks slight weekly support and resistance.
The bullish outlook will be preserved if the pullback stops short of 489. Because OEX 487-488 is such an important level, marking two interim-term tops on the OEX's weekly chart, I would give the OEX leeway to this level. The rising 21-pma currently crosses at 483.57 but might have risen to the 489-490 level by the time the OEX could decline that far.
Although it's not as apparent on this chart as it is on intraday charts, the OEX's two forays toward 505 show up as potential double tops on intraday charts. A move below 489 would technically confirm that double-top formation, predicting another 15-point fall, although I would give the OEX leeway down to that 487-488 level before considering that formation confirmed.
The daily SPX chart appears similar, although subtle differences show up here, too.
Daily Chart of the SPX:
The SPX chart may show more mixed signals than the OEX's. The evening-star formation appears closer to the classic description. RSI turned down more distinctly, although it has not yet and may not violate its rising trendline. Unlike the OEX, the SPX's two tops may have been far enough different in price to disqualify the pattern as a potential double top.
However, ADX measures 39.32, showing an intact trend. The SPX sprang up ahead of its midline support level.
If Monday sees the SPX rise, the 1007.69 June 5 high might be the first place to watch for a pullback. If the SPX instead falls through that midline support, the ascending channel's integrity would be preserved with a fall to a level as low as 965. Currently, the 21-pma rises underneath that channel, crossing now at 958.82 and perhaps rising to meet the SPX at the 965 level if the SPX should test the bottom of the channel. Watch for a downturn in the ADX level and an RSI violation of the rising RSI trendline for first signs that a trend change has occurred.
As we've seen over the course of the last months, sentiment can undo anything shown on the charts. Serious Fed-guessing begins next week and much attention will focus on economic numbers. Looking ahead to the week's economic calendar shows a light day Monday, with only the NY Empire State Index and National Association of Homebuilders June Housing Market Index due, but the housing number might draw some interest.
Tuesday and Thursday's calendars prove full. CPI and housing starts launch Tuesday's releases before the bell, with capacity utilization and industrial production released at 9:15.
Both CPI and housing starts may be watched closely. Worries about a housing bubble keep percolating, and many eyes focus on the housing numbers, watching for the first weakening pinprick. Earlier this week, homebuilder Lennar showed no such weakening. Shares rallied to a new 52-week high after the company announced Q2 earnings of $2.05/share, handily beating the year-ago earnings of $1.37/share and the forecasted earnings for this year. The company also noted a favorable outlook for the homebuilding market, although the CEO did note pockets of regional weakness when he spoke on CNBC.
When discussing retail sales figures this week, one economist pointed to the cold, wet weather in much of the nation as one factor impacting retail sales. The same weather conditions could impact housing starts, with the same reason giving to any disappointment in those figures. In addition, the Mortgage Bankers Association of America released figures showing that the purchase component of its index of applications fell to 418.9 from the previous week's record high of 460.5.
Housing starts might just be lofted by the continued new lows in mortgage rates, however. This week, the average for a 30-year mortgage dropped to 5.21 percent. Perhaps the greatest impact in those lower mortgage rates will occur several months from now, with building permits, at least, usually increasing a few months after a drop in mortgage rates. Freddie Mac's chief economist said that Freddie Mac had raised their forecast for 2003 origination volume. Today, the Dow Jones US Homebuilders Index fell only 0.15 percent, against an $SPX decline of .99 percent and a Wilshire 5000 ($TMW.X) decline of 1.03 percent, showing relative strength as compared to the broader markets.
CPI predicts inflation, or disinflation these days, and will be closely watched for that reason. Economists forecast a decline of 0.1 percent in May CPI, adding to the prior drop of 0.3 percent. Forecasts are for a 0.1 percent fall in the core number, slightly better than the prior 0.3 percent drop. That core number includes breakdowns into apparel, tobacco, airfare, new car, and other components. This week's inventory numbers showed a buildup in inventories of new cars, perhaps predicting lower numbers in that component.
May industrial production forecasts range from 0.0 percent to 0.1 percent growth, better than the prior 0.5 percent decline. Manufacturing production, one component of Tuesday's industrial production number, might be important to watch.
Economists expect May capacity utilization component to measure 74.4 percent, the same as the prior period. Capacity utilization numbers over 85 percent are considered inflationary, but we're not likely to see a number that high in this report.
Thursday sees the Q1 current account figure and initial claims before the bell, May's leading indicators at 10:00, the Philadelphia Fed's release at noon, and the treasury budget at 2:00 ET. The continuing claims number will likely draw the most attention, as this previous week's number showed the four-week average hitting a 20-year high, rising 2,250 to 433,750. This four-week average smoothes out seasonal fluctuations and so is considered most trustworthy. This week, initial claims dropped, with the 430,000 number still higher than expected, remaining above 400,000 for more than four months.
I'm tempted to propose that with markets being so overextended, any number that worsens hopes for economic recovery or else lessens hopes for a rate cut might be the impetus for that pullback we've all been awaiting. However, if Friday's Michigan sentiment numbers weren't going to be that impetus, I don't know what will, and those numbers did not produce significant technical damage to the indices.
This feels as if the markets want to charge right up into the FOMC meeting week after next. On each chart, I've marked the likely levels where pullbacks might begin, and the levels to which the indices could pull back and still preserve bullish sentiment. Moves over those resistance levels might prompt a next leg up to next resistance, while a fall under the marked support levels might portend that the deeper pullbacks will begin.
Don't forget that in addition to the next week being option expiration week, June marks the end of quarter and end of half, too, and some end-of-quarter-and-half positioning might be occurring as we move into the second half of this month. Fed-guessing at full alert, triple expiration, extended markets, and the approach of the end of quarter and end of the half: that sounds like a prediction of volatility to me. Trade carefully.