Meet Ernesto. He’s just a young ‘un, but he is expected to be quite something when he matures. Right now, he is a tropical depression near the Caribbean’s Windward Islands, but according to a forecaster at the Miami-based U.S. National Hurricane Center: “As it moves over the warm waters of the Caribbean, it could get even stronger. Conditions are favorable; it’s just a question of when.”

Ernesto would be the fifth named storm of the current hurricane season, with no hurricanes yet. On average there have been 4.6 tropical storms by this point in the season, but by this time last year, the Atlantic had produced 11 named storms, including four hurricanes. This week marked the one-year anniversary of the 11th storm’s formation, Hurricane Katrina, which caused $81 billion in damages.

For the Equities Portion of Your Portfolio

The hurricane season runs from June 1 to November 30, but the peak season is mid-August through October. The forecasters at Accuweather expect a burst of tropical storms during the next four or five weeks, beginning with Ernesto. It’s another reason (added to Iran, Nigeria and Iraq) to think now is a good time to be buying New Energy Technology stocks.

What you will want to remain cautious about is jumping into full positions in other areas. The market may or may not crack before the end of August, but I still haven’t seen anything to change my outlook for a substantial drop into a mid-October low.

Certainly, two of the stalwart areas of technology are weakening, with a third — business capital spending on software — about to crack. Cell phones have been an area of surprising (to me) strength, but it looks like there’s an inventory adjustment going on there right now. National Semiconductor cut its sales outlook due to lower-than-expected sales to mobile phone customers. Analog Devices said that Asian cell phone companies are cutting orders because they overestimated demand for handsets.

PCs have been an area of surprising (to Wall Street) weakness. As I expected, the market research firms are starting to slash industry sales estimates for this year. In one of the most pathetic marketing balloons I’ve ever seen released, Microsoft is thinking about offering Vista discount coupons to spur hardware sales during the holiday season. That’s ridiculous on at least two counts:

1. The odds that Vista will ship on schedule in January are very small, which means there’s almost certainly going to be another embarrassing delay to turn people off the idea of upgrading now. Microsoft operating system releases are usually delayed several times — the original name for Windows95 was Windows94, and the delay led to a poor year for PC sales. Typically, production versions are released about nine to twelve months after the second beta version. The second beta for Vista was released in May, so that implies that we should expect a production version in the second quarter of 2007 rather than the current expectation of January. But the second beta has been widely reviled as unstable and lacking many promised features, so I think even that is an enthusiastic schedule.
2. Very few people are willing to go through the pain of a Microsoft operating system upgrade to save $30 or so on Vista. People buy new computers with the operating system preloaded just to avoid the upgrade situation.

If this discount program even sees the light of day, I doubt it would change my outlook for PC demand one bit. The Gartner market research firm now says Vista’s delay will cause the PC industry to lose about $4 billion in revenues this year, and they’ve changed their revenue forecast from flat to down 2.3%. That’s around four million PCs that won’t be bought. I think the real number is closer to fifteen million, and industry revenues will be down 10% for the year. Because 30% to 40% of all PCs are bought in the fourth quarter, many for the holidays, we probably won’t see the real shortfall until the year is over. But assuming Vista is delayed to the spring, the PC stocks should move down well before the holidays. I still want to own more semiconductor stocks, Intel, Microsoft and maybe Dell for the Vista MegaShift, but this is not the time to get ahead of the news.

These are some of the reasons why I’ve recommended that you keep a portion of your portfolio in cash. I had a subscriber email from Zack: “Again, you state we should be holding a lot of cash. If that’s the case, why have you consistently recommended that we continue to buy stocks during the April-July stock decline? If you were a realistic advisor, you would have recommended that we sell and wait. Then we would have some cash to invest. You are not consistent in your advice when you advise we buy stocks and hold cash at the same time. You don’t make sense.”

Sorry, Zack, I thought I’d said this enough, but I’ll say it again. Beginning with our timely sale of a large percentage of most subscribers’ portfolios at the beginning of February, followed by an equally timely sale of the China Internet stocks at the end of May, I have consistently said to hold a large percentage of your portfolio in cash — but not 100%. For the portion that remained invested, plus new cash flows into some subscribers’ portfolios, plus new subscribers joining us with nearly all cash (I met some at the Washington, D.C. Money Show), I have stocks on the buy list, Top Buys and new recommendations.

So, Zack, what you need to do is ask yourself five key questions that will help you determine how much cash you should have in your portfolio. The first portfolio decision you need to make is a strategic one: How much cash should I hold?

The second decision is also strategic: Which companies do I want to associate my capital with?

The third decision is tactical: Is the stock attractive at today’s price?

The fourth decision is also tactical: Should I buy a full position, or average in over two or three buys?

The fifth decision is strategic: What should the relative positions be among the equities in my portfolio?

The answers to these questions are as individual as each of you. I can tell you what an “average” portfolio might be, but like the good people of Lake Wobegone, I think all of you are above average. Under SEC regulations, I can’t give you individual portfolio advice, but I try to make it pretty plain what I would be doing with a clean-slate portfolio today. That becomes a model to move your portfolio towards, understanding that there always are real-world constraints, like brokerage commissions and tax considerations.

Now that you know how to determine how much cash to put aside, here are some updates on our companies, which may also be bought under their buy limits should you chose to.

Avian Flu MegaShift

It’s the dog days of summer, with very little going on, which accounts for this shorter-than-usual issue. But the avian flu is about to move back on to the front pages as the southern migration of birds from Canada and Alaska begins. I was showing my daughter how to cast a spinning reel yesterday when the first flock of Canada geese came honking in for a landing. They’ll be here for a couple of months and then head further south. And while they’re in the area, I’ll be making sure that neither they nor the resident Muscovy ducks try to share any of the chicken feed. Ducks don’t get bird flu, but they can carry the H5N1 virus from wild birds to domestic poultry.

Indonesia reported seven sick people in one village, again raising the possibility of human-to-human transmission. One 35-year-old woman died, which was the 95th bird flu case this year, matching all of 2005. She also was the 63rd death in 2006, about a triple from 22 this time last year, and already 50% higher than the 2005 total of 41 deaths. With the seasonal flu season starting soon, it looks like a very bad year.

The Jakarta-based chief of the health and nutrition unit of UNICEF said: “We have a national pandemic among birds and that is where we should be focusing.” The Indonesian government will spend $54.4 million this year to fight avian flu, but due to a lack of resources they have cut next year’s budget 15% to $46.5 million. So much for getting control of the situation.

BioCryst Pharmaceuticals (BCRX) remains my Top Buy up to $19 for a $30 target after H5N1 is found in the U.S. Crucell (CRXL) is a buy up to $28 for my $50 target. Gilead (GILD) has listed January 2009 LEAPs, and I seriously considered advising those holding the January 2008 $50 calls (YGDAJ) to roll out to the new January 2009 $60 call (OJKAL) and take some profits off the table. However, I think we should stick with the ’08s, which have a lot of leverage and may give us long-term capital gains treatment. I would still buy the January 2007 $60 LEAP (GDQAL) under $9 for a $20 target.

Biotech MegaShift

Biogen (BIIB) has been trading sideways for nine months, and the time premium on the January 2007 $45 LEAPS (IDK AI) is burning off. When the stock went under $45, the options dropped sharply, and I received several subscriber emails from Greg and others questioning the $21 target and asking if it is smarter to average down in the 2007s or roll them out to the 2008 $45 LEAPS (YZUAI).

I understand your concern, but there are two reasons that I think my $21 target is still good, which implies a $66 price for BIIB when the options expire in January.

• I’m expecting a sharp market rally from mid-October into January, with biotech outperforming in the downturn into mid-October and then being the #1 or #2 industry group (with semiconductors) in the rally.
• I’m expecting Biogen to report surprisingly strong Tysabri sales as MS patients worldwide clamor to get the drug.

As for averaging down or buying the 2008s, I am always in favor of diversifying. The 2008s give you time diversification. I would not roll the 2007s into the 2008s, as I think the risk/reward is similar when comparing the two options. But I would put new money into the 2008s rather than average down in the 2007s.

Dendreon (DNDN) submitted two sections of their rolling BLA (Biologics Licensing Application) to the FDA, covering their clinical and non-clinical testing for Provenge. The clinical data shows a survival advantage without any significant toxicity. They will submit the chemistry, manufacturing and control sections as the year rolls on, which will complete the application before the end of the year.

The rolling filing process is part of Fast Track Review, which should get them an FDA decision by mid-2007. Provenge will be the first therapeutic cancer vaccine, and the exact same process can be applied to other solid tumor cancers like breast, head and neck and melanoma.

However, the FDA could always turn Provenge down — I’ve seen them do things that dumb before — but extending survival is the gold standard for a cancer drug, and without any toxicity, it is difficult to see why the FDA might nix it. Clearly, Wall Street believes Provenge will not be approved, probably because the clinical trials missed their primary endpoint of tumor progression. Survival was a secondary endpoint. I think the Street is wrong, and I would continue to buy DNDN under $7 with a $14 target after Provenge is approved.

Content on Demand MegaShift

Harmonic (HLIT) is buying a venture-backed private company, Entone, for $45 million — $26 million in cash and about 3.54 million HLIT shares. Entone makes software that turns an ordinary server, usually from H-P or Dell, into a Video on Demand (VoD) server. It is a direct competitor to Arroyo, which was acquired by Cisco for $92 million on Tuesday. These two companies are the leaders in VoD software systems, with Arroyo mostly targeting cable companies and Entone mostly targeting telecom and satellite companies. Entone also has close ties to Tandberg, a main Harmonic competitor. Tandberg even invested in Entone, so this could be interesting.

This acquisition hardly budged HLIT’s stock price, but it is a big deal for a number of reasons. First, it will be accretive to earnings in the first quarter of 2007, which is really quick and suggests Entone is growing very rapidly. Harmonic paid half as much for Entone as Cisco paid for Arroyo for roughly the same $10 million revenue run rate, which is part of the reason it will be accretive so quickly. About 50 of Entone’s 80 employees will join Harmonic’s team, mostly R&D engineers based in Hong Kong.

There are a number of companies, like Seachange (SEAC), that produce expensive video servers based on custom semiconductor designs. Entone takes advantage of the ever-cheaper standard servers and can more easily add features in the future because it is software-based rather than hardware-based. Harmonic now has an end-to-end VoD solution without having to add a custom video server product.

Entone has 35 customers, including Hong Kong-based PCCW, which is the largest IPTV supplier in the world and is supplying HDTV over DSL lines. That installation uses Tandberg TV’s encoders instead of Harmonic’s. Harmonic could continue to support Tandberg, or (more likely) choose to migrate the Entone software away from that competitor to tie it to Harmonic encoder sales. Entone is the #1 VoD server supplier to smaller telecom companies through an extensive distribution network, and that is a market Harmonic has not had the time and resources to address in the past.

The Content on Demand MegaShift that we’ve identified is going to happen very quickly. Most of us will watch VoD within five years, and the idea of a regular broadcast schedule probably will be dead in ten years. This was a superb acquisition, and HLIT remains a Top Buy under $6 for my $12 target.

Telkonet (TKO) rallied on the EDS contract announcement, but then gave back about half the gain. That is typical market behavior, and I’d like to see the stock base around $2.50 before the next leg up.

Dick asked: “Are the announcements about TKO and EDS the end of the good news until their earnings?”

No. There will be a large investment by a strategic investor that will give TKO a lot of credibility and attention, and they will be appointing a new CFO. There also should be a series of order and contract announcements as Broadband-over-Powerlines hits the growth curve. TKO is a great buy all the way up to $5, and I am sticking to my $15 target (a $750 million market cap) as the aggressive small cap growth funds discover it in the coming bull market.

Security MegaShift

Symantec (SYMC) quietly increased renewal pricing for its 2006 consumer products like Norton Internet Security. In late 2005 they introduced autorenewal. The combination of autorenewal and higher prices is likely to provide some upside surprises in the September and, especially, December quarters. SYMC is a hold for a higher exit price in the fall.

WiMAX MegaShift

I said last week that the more I looked at Sprint’s $3 billion commitment to WiMAX, the more I liked it. It’s going to change the way people access the Internet, cut costs and provide fourth generation (4G) mobile phone service. The higher speed, longer distance features of WiMAX work best in the licensed spectrum, although WiMAX also has several unlicensed spectrums for the standard. Sprint dropped $1 billion in the early ’90s to buy spectrum for a Multipoint Multichannel Distribution Service (wireless cable TV), which never went anywhere. That same spectrum can be used for WiMAX. Then, when Sprint bought Nextel, they acquired even more of the spectrum that they need to convert to a nationwide WiMAX system. They already have the Nextel and Sprint towers in place, so they don’t have to buy spectrum or lease towers — they can just deploy the equipment. Sprint’s core cellular network is based on a low-power technology called PCS (Personal Communications System) that required lots and lots of towers. That’s a tremendous competitive advantage for a WiMAX network, and explains why Sprint is so ready to spend seriously big bucks on this project. It’s also why they jumped straight to mobile WiMAX — they can upsell their 51 million mobile customers, and provide fixed service along with the mobile. Cell phone competitors are afraid to sell their data services for fixed applications, because they might get overwhelmed by the demand for bandwidth. WiMAX has plenty of bandwidth for all — voice, video and data.

Airspan (AIRN) said Last Mile Broadband, an Irish Midlands broadband supplier that put 50 Airspan base stations in place in 2005, just ordered another 35 to launch higher speed broadband and VoIP services.

Michael, a subscriber, wrote that he read the 10Q for the quarter, but apart from the revenues being significantly higher than the consensus, there was no real reason for the negative price action. He asked: “What happened?”

Michael, I’ve covered this in previous Radar Reports. Their big Japanese contract with Yozan is in turmoil, both because it isn’t clear how much more product Yozan will take and due to very technical accounting issues that led them to book a lot of deferred revenue in the recent quarter, while leaving it very unclear what they can book for the rest of the year.

The key to a stock like AIRN is focusing on the forest — is WiMAX growth accelerating? And then the trees — are they getting their share of orders? So far, the answers are yes and yes. Twelve months from now, Yozan and the accounting issues will be history, WiMAX will be one of the very hot stories, and Airspan, Alvarion and Terabeam (TRBM) will be on everyone’s watch list. That’s why I want you to keep buying AIRN under $6 for my $10 target. Incidentally, I don’t think it has any meaningful risk from current levels.

Alvarion (ALVR) has also won a couple of big contracts. Netia, one of the leading communications providers in Poland, chose Alvarion’s BreezeMAX for a 20-city WiMAX deployment to be turned on at the end of August. That’s an amazingly fast deployment time, and a tribute to Alvarion’s no-hassle solution. Three other companies in Poland were granted WiMAX licenses, and it will be interesting to see if Alvarion can get some of the other business.

A subsidiary of NTT, the Japanese telephone company, choose Alvarion’s BreezeACCESS VL system to bring broadband to Okinawa. A wired infrastructure would be too difficult and costly, due to the mountains and rural character of much of the island. ALVR remains a buy under $9 for my $18 target.

MobilePro (MOBL) won another plum Wi-Fi contract for Longmont, Colorado, which has one of the highest concentrations of software jobs in the nation. They’ll start deployment in the fourth quarter, and as usual, they will give the city free access for government services. The network will cover 22 square miles and more than 71,000 residents.

The PowerPoint presentation MobilePro management gave at the annual meeting is at http://www.hawkassociates.com/mobilepro/company.php, where you can also listen to a replay of the talk. They said that after the Sacramento experience, they will not make a press release on new wins like Longmont until they have a signed agreement. They used to announce a win when the city council voted for it, but that put them at a negotiating disadvantage. Over 700 cities have indicated that they are going to look at building a municipal Wi-Fi network. MobilePro needs $18 million in capital to build out these networks over the next 18 months, and they want to pay off the $15 million convertible in cash. They have $9 million in funding that will close in the next couple of weeks, plus $15 million in serious discussions, and another $10 million in a terms sheet from an international fund, on which they are doing due diligence. They are looking for equipment financing from Cisco and Motorola, their two largest suppliers.

MOBL said that they have a globally known advertising partner announcement coming this fall for the WazTempe network — it could be Yahoo. They also said that they are targeting a stock listing, either by a reverse merger or a reverse split. The CEO said they are in preliminary talks for a reverse merger, and he personally is not in favor of a reverse split. He is targeting a listing by the end of March, and part of his bonus is tied to that. They are very disappointed with the stock’s price, because they feel they have addressed the major concerns of organic growth, funding and are working on a listing. They emphatically said that they are trying to build long-term value and plan to be here for years.

David, a subscriber, wrote to point out that I said I would report on my review of the company’s 10Q, and then forgot to do it last week. He’s right. There was nothing unusual in the filing, but I should have told you so. Buy MOBL up to 25 cents for a 60-cent target.

Google Short

Google (GOOG) stopped us out at $385 last week and almost immediately fell back. I expect it to try to rally a bit here and then fall away.

In the last 18 months, insider selling at the top of Google totals over $7.4 billion. Here are the big chunks:

Seller	Amount
Sergey Brin:	$1.9 billion
Omid Kordestani:	$1.1 billion
Eric Schmidt:	$650 million
Ram Shriram:	$650 million (Director, Investor)
David Drummond:	$200 million
George Reyes:	$200 million
Jonathan Rosenberg:	$200 million

Needless to add, not one of them has bought a single share of GOOG on the open market. Maybe they see the advertising market weakening in an economic slowdown, as always happens. In any case, short GOOG over $370 with a $384 stop loss on a closing basis. My target is still $200, and the December $380 puts (GOP XP) are only a little overpriced relative to their theoretical value.

Market Outlook

I used to feel like an alarmist about the housing bust, but now that it is headline news, I’m not so alone. July sales of previously-owned homes in the U.S. were the lowest in two years at an annual rate of 6.33 million. That was lower than the lowest forecast, let alone the consensus. Sales are down 11.2% from last year, while the inventory of houses for sale exploded to a new record — 3.86 million, or a 7.3 months’ supply.

July new home sales, announced this morning, fell more than forecast to an annual rate of 1.072 million from 1.131 million in June. Sales were down 22% from last year. The inventory of unsold new houses rose to 568,000, or a 6.5-month supply — the highest level in 11 years. Purchases of used homes fell in all geographical areas, while new homes fell in the Midwest and South, so this is not a bicoastal bubble bursting — it is a serious recession in this sector.

The median price of an existing house rose 0.9% in July from a year ago, to a record $230,000. New house prices rose 0.3%, also to a record $230,000. With sales down in double digits year-over-year, it seems likely that prices will turn down. Potential buyers, well aware of the inventory situation, say they are waiting for prices to fall. I think this housing recession will extend well into 2007, acting as a significant depressant on economic growth.

The bond market seems to agree. The yield curve is still inverted (lower yields on 10-year notes than 2-year notes), which means the bond vigilantes expect a big slowdown. But at the same time, the spread between Treasury Inflation Protection Securities (TIPS) and regular Treasuries is getting wider and wider. That means the bond market also expects more inflation. Hmmm, a slowing economy with rising inflation — S-T-A-G-F-L-A-T-I-O-N anyone? Higher inflation will force the Fed to raise rates again, even with growth slowing. When the stock market figures it out, which I expect to happen by mid-October, we should see the bottom created.

 Print This Post