Avalon Asset Management Company Lexington, Massachusetts Investment Management
Avalon Asset Management Company Surfin' USA
“We'll all be gone for the summer, we're on safari to stay.” – Brian Wilson (The Beach Boys), Surfin' USA
It was another Beach Boys kind of week, as avid traders waxed down their surfboards and rode the waves of good feelings that a summer rally brings. Revenues may be falling, along with employment and income, but bullishness is surging. What matters most in a momentum market isn’t real earnings, the real economy or even the real world. It’s about expectations, baby, and the market is surfing them beautifully.
Airline traffic plunging? Hey, just another example of the savvy cost-cutting that’s producing all those profits. Air pocket in Chinese stocks? No worries, the government won’t let anything bad happen. Worst recession since World War II? Most excellent, because all those inventories have to be replenished now. Accounting quirks producing most of the profits? Well, it’s about time the system worked to our advantage. Surf’s up!
Inventories are indeed low, but to quote CNBC’s affable Bob Posani, “folks, we already know that.” Spending is low too, and while a boost in auto production is widely expected to lead to some production increases, we are sticking with the same outlook we’ve had since the beginning of the year: stuff runs out, so an inventory restock is inevitable, we have to get at least a bounce. So far, though, it still looks like it’s only going to be a bounce. The inventory-to-sales ratio remains quite high.
At one point during the week, Pimco (the country’s best bond manager) co-head Mohammed El-Erian was seen saying that the equity markets were “on a sugar high.” He went on to observe that cost-cutting is not a sustainable source of profits, that the stimulus will run its course, that the flattening in home prices is a necessary but not sufficient condition (and something of a comparison illusion anyway), and that the market’s assumption of a cyclical fall followed by a cyclical recovery isn’t going to happen, despite people’s expectations.
One reason for that might be that extraordinarily tight credit conditions aren’t letting up for anyone but large corporations, and that is primarily due to the demand from yield-hungry investors in this zero-interest-rate world. Yet despite all of this, belligerent bulls were defiantly declaring by week’s end that “a sugar high is just fine.”
As we make clear below in this week’s Economic Beat, last week saw a stunning display of misread, misreported and misunderstood economic data. Wall Street clearly doesn’t want to let go of its momentum market. Some commentators are sick of the recession and others seem to be trying to talk us out of it. Bullish sentiment is rapidly soaring in the markets. The S&P has rallied 46% from its low this year – the sharpest such rally since the Great Depression – and the 1000 level will probably be tested on Monday. We’re just about to the top of the wave.
Two years ago last May, we wrote about the difficulties in getting the herd to change direction. The first potholes are ignored, as are the first ones to go over the cliff. Last week the Chinese market had a sudden air pocket, and it's clear that the government is going to pull back on the bubble lending that has so excited the rest of the world. The Fed warned last week that commercial real estate is going to get seriously worse. Airline traffic – both passenger and cargo – is falling off a cliff, but it’s blamed (by the Street) on fears about swine flu (can cargo get swine flu?).
In short, the picture of most of the globe is fading demand, non-flowing credit and rising unemployment. It’s no longer the abyss, so we seemed to have latched onto the notion that the rebound is without limits. Analysts are back to marking their opinions to the stock price, and twelve-month stock targets are expected to fall within twelve trading days.
The market is now about ten to fifteen percent higher than fair value. It can go higher still and probably will, but the storm signs are in the air. What was the fatal flaw in this decade’s real estate market? The same as the flaw in the tech wreck, namely the assumption that prices will always go up. The amount of optimism in the air and the explanations for it are remarkably similar to this time a year ago, and this time two years ago.
Just remember one thing: we won’t get a big downdraft just because somebody drops an earnings bomb, or we get a reversal in the jobs report. What’s necessary is that we have to get everybody leaning the same way first. The riptide is most dangerous when everybody knows all about it, the weather is fine and nobody is afraid to paddle out anymore. It’s the ones who are sure they can stick around for the last wave that end up disappearing.
Mark Twain popularized the notion that lies consist of three types: “lies, damn lies and statistics.” We have written before that he might well have added a fourth category of government statistics, but that may be unfair. It may simply be that the greatest deceptions are practiced with government statistics, with the media leading the way in its twin starring roles of greatest dupes and greatest dupers.
We could hardly believe some of the nonsense from all corners of the media last week. CNBC announced that new-home sales had “skyrocketed,” alongside a flashing box of “Dow Breaks 9100.” That approach was largely echoed in The Financial Times and repeated in news outlets throughout the week.
In the wake of the report on second-quarter GDP, RDQ Economics was quoted in the Wall Street Journal as saying that that “trade added to growth,” which was like saying that the bubonic plague had improved the food supply outlook. A National Public Radio announcer earnestly suggested that consumer confidence would grow upon the news of the GDP number, simultaneously ignoring the reality that the average consumer hardly knows where current GDP stands, the distortions in the result (in dollar terms, GDP actually fell more than expected), and the drop in consumer confidence that was reported only three days earlier.
We may as well begin our dissection with the new-home sales report, a bit of data that ended up sounding like the North Koreans had written it. For a start, new-home sales totals are running at such low levels that any changes translate into much larger percentage moves, a feature of much economic data these days.
The actual reported data for the month was for 36,000 new homes in June versus 33,000 in May. That’s not a good number, but a terrible one. In forty-five years of data, the only June with lower sales volume was in 1982. In that year, the population of the country was two-thirds of what it is now, and there wasn’t any $8,000 tax credit. On a per-capita or any other relative basis, this is unquestionably the worst June since records began in 1963.
What’s more, a very small revision, say from 36,000 to 35,000, would cut that biggest-percentage-increase-in-years nearly in half. Such a revision is hardly far-fetched, as April now totals 14,000 less than first reported (annualized) and March 24,000. The title of the lowest June ever reported is still within reach, requiring a reduction of a little more than 1000 homes to the monthly total.
Another interesting feature of the data is that the number of homes for sale at the end of the period fell sharply. That suggests that (a) there was something of a clearance sale going on; (b) we have probably peaked for the season’s sales rate (June usually marks the peak); and perhaps most importantly (c) a potential drop in July sales looms. There could be some real disappointment coming.
It’s important to understand the seasonal adjustments. June is supposed to be the peak month for actual new-home sales, which generally decline into July and then rise again in August. These are only model estimations, of course, and actual sales in the summer don’t always follow the typical pattern, nor do the variations tell the same story.
In the double-dip recession year of 1981, for example, actual sales of 38,000 in June fell to 36,000 in July. This was reported as an increase of 3.9% in the seasonally adjusted annual rate (the headline number), because July sales were somewhat higher than what the model called for. The June number, however, turned out to be a better representation of the real trend, which is about what you’d expect from what is usually the peak month.
The usual pattern for actual new-home sales is to peak in June and be somewhat lower by the end of the summer. That’s only the typical pattern, though, and even banal years don’t always fall neatly into that slot. It isn’t that unusual for summertime sales to be roughly level for the three months, which causes big variations in the headline number. If July is about the same as June, it gets reported as a big increase, but then if August is about the same as July, it gets reported as a big decline. Be careful about monthly data.
The best month to represent the underlying trend is indeed June. Its annual rate is usually within a few percent of the year’s actual total, and nearly always within five percent. Current weekly real-time data continue to show extreme weakness in the mortgage-purchase market, declining retail spending, and rising unemployment. We don’t expect banks to start lending again this year, so our conclusion is that the revised June number will be the best indicator of this year’s total. That means the worst year for new-home sales since the data started being kept in 1963, and a bottom that lasts into next year.
A year ago we complained that the GDP price deflator for the second quarter looked improbably low at 1.1%, and said that the quarter’s reported GDP (1.9%) was unlikely to be accurate. A year later, the reported price deflator for Q2 2008 has jumped to 1.7% while revised GDP has fallen to 1.5%. This year the current price deflator tumbled again, to only 0.2% versus the first quarter’s 1.9%, while the deflator for gross domestic purchases actually rose to 0.7%. We smell another rat.
In fact, we smell a roomful of rats with the advance estimate of second-quarter GDP, whose reported annualized rate of (-1.0)% was hailed as some kind of victory by the momentum bulls and imagined to inspire new confidence in the consumer. The consensus, you see, called for a decline of (-1.5)%.
However, that consensus estimate was based upon the first quarter’s GDP, which in its “final” estimate of a month ago was supposed to have fallen 5.5%. It’s difficult to wade through the Bureau of Economic Analysis (BEA) numbers for direct comparisons, because the benchmarking numbers were given an annual revision in July and on top of that, the BEA switched its base for real dollar terms from 2000 to 2005.
In spite of these complications, though, we can tell you that using dollar totals, second-quarter GDP fell at a (-2.0)% annualized rate when one uses the original “final” estimate for the first quarter. The notion that things are somehow better than expected is further belied by the fact that first-half GDP is now reported to have fallen at an annual rate of (-3.8)%. It would be at (-3.5)% if consensus had been correct. The year-on-year change widened from (-3.3)% in the first quarter to (-3.9)% in the second. According to the data, the economy is doing worse than estimated, not better.
If all of this is giving you a headache, we don’t blame you. We said last week and many times before that the GDP headline totals aren’t as important as the market makes them out to be. They are often quite poor representations of real economic strength and are heavily revised to boot.
For an example of the former, consider the benefit to the number from the collapse in trade, reported by MarketWatch.com as, “the benefit from trade”(!). As the global economy continued to weaken, exports and imports fell. As Americans cut back on spending, imports have been falling faster than exports. This contributed to a whopping 2.14 percentage points in the quarterly GDP measure. In other words, had we been willing and able to buy as many imports in the second quarter as we did in the first, annualized second-quarter GDP would have fallen by over three percent, not one. Some strength.
To top it off, there are the revisions. The latest revised data for quarterly GDP in recent years are dramatically different from previous reports. 2008 GDP was revised from 1.1% to 0.4%. The decline from the fourth quarter of 2007 to the first quarter of 2009 was given a full extra percentage point from (-1.8)% to (-2.8)% annualized. The fourth quarter of 2007 was originally reported to be (-0.2)%, but now we are told it’s plus 2.1%. Over long periods, these revisions tend to blend together and leave the running total intact, but the quarterly numbers are notoriously unstable. The market trades on them anyway because they represent news, but they aren’t much more than noise.
As to the rest of the week, as noted above, the Conference Board reported that consumer confidence fell unexpectedly for the second month in a row, from 49.3 to 46.6. The confidence numbers often don’t predict spending very well, but for what it’s worth, consumer attitudes towards the job market deteriorated, and fewer respondents indicated any plans to purchase a car or home.
The Case-Shiller report was widely cited to show “surprising” strength, as many cities recorded small monthly gains and the year-on-year rate declined. We expect that the year-on-year declines will continue to improve, not because the market itself is improving, but because the comparisons will get easier as we approach last year’s second-half abyss.
We also believe that as the foreclosure mix changes, it will introduce positive variations in the data; they may later be followed by negative ones if an anticipated rebound in foreclosures materializes. In addition, the mix should change as inventory in the lowest quartile, which is the most bombed-out category, gets cleaned out. This may benefit pricing data.
We will also repeat our observation that over the last year or so, changes in the home buying environment have brought in waves of buyers that have been consistently smaller than the market wanted to believe. Rather than represent a trend, the pool of incremental buyers able to respond has been limited, and the surge has ended in a month or two. Last December’s drop in mortgage rates is a good example. In the current situation, the summer buying season and the impending expiration of the tax credit for first-time buyers may be repeating the pattern. Credit is still extraordinarily tight.
New orders for durable goods fell more than expected. It’s a volatile category, and if you wanted to be an optimist (or are fully invested) you would say that excluding transportation, new orders rose 1.1%. However, if you are a realist, and we try to be one, you would note that excluding defense, new orders fell 0.7% and shipments fell again for the eleventh month in a row.
In a good example of our thesis that we have to get sideways sometime soon, the category for business investment showed a monthly increase of 1.4%, while the year-on-year total is down nearly twenty-four percent. We would also add that all of the May data was revised downwards, which means that the dollar totals were a bit worse than the percentages would lead you to believe. Factory orders, to be released next Wednesday, will include non-durables as well and are expected to run about the same way.
In another sideways story, the Chicago PMI (Purchasing Managers’ Index) report for July improved to 43.4 from 39.9, while the new orders category “leaped” to 48.0. What this means is that conditions declined for the tenth month in a row, but that the rate of decline softened again. The market may get excited when we finally get to a neutral monthly comparison, but from our point of view the time that it’s taken just to get to one neutral month is pretty ominous. Still, Chicago should catch a break soon from expected restarts in the auto industry.
The national (a.k.a. the ISM) number for manufacturing is due on Monday at 10 A.M., and the cautious consensus is for more of the same: another decline, but a smaller one creeping back towards neutral. Every month seems to set a new all-time low for capacity utilization, meaning we’re very overdue for an uptick. Construction spending for June comes out at the same time, with the usual expectations of a small decline leaving the usual door open for any kind of positive surprise. It’s the first day of the month, so traders will be inclined towards benevolent interpretations.
We told you about weekly claims; after a strong uptick last week they are expected to decline slightly. We expect continuing claims to continue to diminish as time runs out on the unemployed, but the total unemployed number will get a better look of sorts when the monthly jobs report comes out on Friday. The losses are expected to slow to 300,000, while the unemployment rate is expected to continue climbing towards 9.7%.
The jobs report is another volatile series that is prone to heavy revisions, and we’ve been making the case for some time that the Labor Department’s adjustment models don’t do well in the kinds of conditions that we’ve been experiencing. The consumer confidence report from the Conference Board last week points towards further losses, but what gets reported could be something quite different. The auto industry has turned the department’s weekly claims numbers inside out, putting a wide miss on the unemployment rate into the mix. We could get a real head fake.
Let’s see, with private wages showing the smallest yearly increase since 1980, continuing claims expiring and unemployment rising, what do you think the impact on spending will be? The market thinks it’s very little, with expectations of a spending increase of 0.3% in June to be reported on Tuesday. Personal income is thought to have fallen back along with government payments.
The ISM non-manufacturing index is also expected to creep back towards neutral when it reports on Wednesday. Last Wednesday the Fed released its Beige Book regional survey, which reported a cautious mix of slowing declines and stabilization. Notwithstanding, we think that the continued weakness in transportation is another canary falling over in the coal mine, although this market hasn’t had time to dwell on problems.
Rounding things out, we’ll get pending home sales on Tuesday, various private-sector jobs reports on Wednesday and Thursday, and a report on consumer credit on Friday. The last is expected to still be falling, along with personal income and unemployment compensation. Yes, time to get the surfboards out and catch that wave of spending.
StockWatcher will return next week.
Avalon Asset Management Company is a Registered Investment Adviser
Avalon's MarketWeek is not intended as a market timing newsletter or service. No buy or sell recommendations are made for any of the individual stocks mentioned on the site, and neither Avalon Asset Management Company nor its officers, directors or employees make public stock recommendations. Please address comments to MarketWeek@AvalonAssetMgmt.com