The market loves bad news again

Lackluster employment growth, but year-over-year growth still constant
The delayed government employment report showed the economy added only 148,000 jobs in September, which is well below the consensus of 185,000, but not that far below my 160,000 estimate.

State and local governments had two great months of growth that disguised a relatively sharp one-month decline in private-sector job growth to just 126,000 jobs added, compared with its 12-month average of 185,000 jobs. The strong growth in purchasing manager data from the ISM for both the composite and the employment sector appear to be mirages, as the economy added just 1,000 manufacturing jobs in September. The other reason for overly optimistic forecasts for the September jobs report were tumbling initial claims reported for several weeks of September. The low September claims reports, in retrospect, were caused by California processing issues and not a real decline in claims.

In any case, the year-over-year private sector employment growth rate remained stuck at 2%. The world isn't falling apart, but it's not accelerating, either. Interestingly, last September was also soft, which in turn helped keep the growth rate stable. Apparently the statisticians haven't caught up with some new seasonal effect that is hurting hiring in September.

Data on hours and wages were also relatively steady and consistent with recent averages. Hours worked were largely unchanged both year over year and month to month, as they often are at this stage of a recovery. Wages were up a relatively weak 0.1% for the month but still managed a 2% gain year over year. Combining all three components, overall wage growth looks to be 4.4%, its best performance of the year.

Existing-home sales back off their torrid pace
July and August both saw a huge surge in existing-home sales as buyers rushed to close sales before rates soared even further. July and August were reported as 5.38 million unit sales, although the August figure was revised sharply downward from 5.48 million. As expected, housing activity slowed down (even when seasonally adjusted) sharply to 5.29 million units in September. Outside of July and August, that is still better than any other month of 2013. The year-over-year averaged data backed off its July highs but still looks OK, but certainly no signs of more acceleration.

While the 13.1% unit growth is exciting enough, the total transaction dollars calculation is even better with 24% year-over-year growth. However, the transaction data is showing some signs of plateauing. Unfortunately, pending home sales data suggest more deterioration next month. Pending home sales, a usually reliable indicator of existing-home sales, have been falling for some time. A sales rate of 5.1 million or 5.2 million for October should come as no real surprise. A drop of this magnitude shouldn't have a big impact on overall economic growth. Slim inventories will weigh on future growth, too. The good news in the report was that inventories are not decreasing anymore on a year-over-year basis. At 2.2 million units of inventory, housing inventory has been cut almost in half, which explains the strong home price data.

Although interest rates have come back some, they still remain elevated compared with a year ago. Those higher rates, combined with the price increases noted above, will reduce the affordability of housing. This lower affordability will hold back existing-home prices for the next several months. I surmise that we have probably seen the best existing home sales number we are going to see until sometime next spring.

Home prices continue to show signs of plateauing
Higher interest rates and declining affordability are also beginning to weigh on home price growth. From July to August prices grew just 0.3%, the lowest rate of growth since September 2012. However, the year-over-year growth continued to move ahead, showing 8.5% growth year-over-year (single point, not averaged). Typical of the last few reports, the data showed double-digit growth in the Mountain and Pacific regions while growth was at a more pedestrian mid-single-digit growth rate for the other seven regions of the country. In fact, those seven regions all fell in a range of 4%-6%.

With the latest data, home prices are now down just under 10% compared with the peak. On average, only buyers purchasing homes between April 2005 and mid-2008 have homes that are worth less than the original purchase price. Still, given that those were some relatively high transaction volume years, that could represent 20%-25% of all current households.

Durable goods orders reports manage to disappoint again
Although durable goods orders looked strong on the surface at 3.7% monthly growth and 9.2% year-over-year growth, almost all of that growth came from the volatile aircraft sector. Excluding transportation, orders for September were down 0.1%, the third month of decline in a row. About the only good news is that the rate of monthly decline has slowed over the third quarter. With another dismal report, the relatively cheery ISM purchasing manager reports are looking a bit silly.

I am never a fan of the durable goods orders report because of issues with airline orders (which are irrelevant to economic activity, because   Boeing  BA has a decade-long backlog even if it never got another order. Aircraft production will not be dictated by orders at all, but by Boeing's production schedule). However, the non-defense capital goods orders portion of the report, excluding aircraft, is a decent barometer of business confidence. Looking at this slice of the data, orders slipped 1.1% month to month, with declines in two of the last three reports and only a tiny gain for the other month. The year-over-year version of the non-defense orders looks better than the monthly data, with data edging up since May. However, the data is just a little suspect because of an 8% jump in January that is skewing my normal preferred methodology.

However, it still seems clear to me that business spending will continue to be lethargic into the fourth quarter. Unfortunately, if businesses don't have enough confidence to buy equipment, it doesn't seem like they will be in the mood to hire a lot of new employees, either.

Shipments of non-defense capital goods are also part of the GDP calculation. The news here is not good. Comparing third-quarter shipments with second-quarter, shipments slipped 2.9% on an annualized basis. So that even after inflation and other adjustments, this is likely to be a subtraction from potential GDP growth in the third quarter.

Trade data shows decent U.S. exports but not much import growth
Both adjusted for inflation and not adjusted for inflation, the trade deficit was basically unchanged between September and August. The data show respectable export growth but lethargic import growth, not good news for our trading partners.

Of course, rising oil- and petroleum-related exports and decreased oil imports are responsible for half the changes. But even excluding oil, exports outshone imports. Based on the last three months of data, net trade will have little effect either way on GDP with real net exports close to identical in both the second and third quarters, at least according to the preliminary data and our estimates.

The long-term improvement in trade data is mind-boggling
Although this month's data is frankly, boring, the longer-term picture is absolutely stunning. The monthly trade deficit in goods and service was a minuscule $2 billion or so when the data series started in 1992, before expanding in an almost straight line to almost $70 billion in 2007. Today that deficit is about half that high, at about $38 billion, with more improvements likely down the road as oil and gas exports continue to expand.

The improvement would look even more dramatic if compared with the ever-rising GDP figures. Many, including me, believe that the United States' huge trade deficit was the root cause of abnormally low interest rates and the housing bubble it engendered. The flood of dollars flowing out of the United States landed mostly back in the U.S.--in U.S. debt securities, depressing rates despite the Fed's best efforts to raise them. The flow of cash and low rates were a powerful combination that led to careless lending and strong consumer demand in the housing market.

Initial claims data: much ado about nothing
Initial unemployment claims remained unusually high this week at 350,000, down from last week's 362,000. This marks the third week in a row of claims above 350,000, which is normally worrisome. Wacky data from California is usually the No. 1 suspect when initial unemployment claims act oddly. This time is no exception with new processing systems implemented on Sept. 1 delaying claims in September, depressing September claims, and then boosting October claims when full-scale processing resumed in October.

Below is the national data, excluding California. I caution that tossing out California is generally not a good idea because it accounts for 15%-20% of all claims nationwide--and it often leads the U.S. economy in both directions. Still, I couldn't resist the temptation to at least look at the data ex-California.

This data shows that claims haven't materially improved since July. And even excluding California, the trend seems to be worsening slightly. Unfortunately, the state-level data is one week behind the national data, so it's impossible to tell if the continued high readings for the week ending October were due to California or not. Stay tuned.

Both new and delayed economic data to flood the markets next week
Next week is already a normally busy week for data, and it will be even busier than normal as the delayed industrial production, retail sales, and consumer price reports will all be released. They join the normally released pending home sales report, Case-Shiller Home Price Index, construction spending report, and motor vehicle sales. Unfortunately, new home sales, housing starts, and permits will be combined for September and October and won't be released until late November or early December. Given the last data we have is for August and was released in mid-September, the housing market will be difficult to forecast or, for that matter, to even determine current conditions.

As usual, I will be watching the auto sales report most closely. Auto sales are a great marker of real consumer confidence. Furthermore, auto sales have been the key driver of this recovery, and any weakness would be a bad thing for the economy. Good news on auto sales would indicate that consumers were brushing off worries about the U.S. budget and debt ceiling debates.

Some odd holiday and weekend calendar flukes made auto sales nearly impossible to interpret. Auto sales for July, August, and September were 15.8 million units (seasonally adjusted, annual rate), 16.1 million, and 15.3 million, respectively. Labour Day sales were included in the August data (normally, Labour Day is included in the September data), perhaps merely shifting sales from one period to the other. The October data will be a bit of a tie breaker. Anything less than 15.7 million units would point to a softening auto market, while anything higher than that would indicate an accelerating market. The consensus forecast is right on the cusp at 15.6 million units. The trade rags are in a tight forecast range of 15.4 million-15.5 million units.

Don't be fooled by the press reports (I've already seen some) trumpeting double-digit year-over-year sales growth. I love year-over-year data more than most folks, but I would not stoop so low as to use Hurricane Sandy-addled sales as any type of benchmark.

Utility data could inflate industrial production, pure manufacturing likely flat to down
Given the relatively strong ISM data, one would assume industrial production would be strong, but that just has not been the case. Industry mix shifts have recently distorted the ISM data to the point where the correlation between the ISM Purchasing Managers Composite and Industrial Production has been zero for the last four years. Consensus is expecting a 0.4% increase in industrial production, but a hefty part of the gain is related to utility output due to cooler weather. Flat manufacturing employment and a slight decrease in manufacturing hours worked suggest that the pure manufacturing sector was flat at best during October.

Meanwhile, the ISM data has risen in a straight line from 49.0 in May to 56.2 in September. The competing Markit reading has been less ebullient and has fallen from 53.7 in July to 51.1 for the flash October reading. The ISM October reading is due next Friday. Consensus is for a drop from 56.2 to 55.1. I would expect something worse.

Auto sales slippage could push September retail sales into negative territory
Retail sales have not been lighting the world on fire lately with very modest gains, especially when excluding autos. However, the low September auto sales report described above is expected to weigh heavily on the headline number, which is expected to show a 0.1% decline. Ex-auto sales, the consensus is for a 0.4% month-to-month gain, which feels too high to me because of lower gasoline prices and soft retail sales data from the International Council of Shopping Centers.

CPI: more good news?
Consumers react strongly to prices in both directions, so low prices often stimulate demand, aiding the economy. Consensus is for a 0.2% increase in overall prices. Given flat gasoline prices, I suspect even that estimate could prove a bit high, with the potential of a headline figure of 0.1% inflation, which translates into 1.6% year-over-year inflation when averaged over three months. Even if the consensus proves closer to the mark, this would be another great report for consumers. Modestly rising inflation would probably be the best reason for more optimism for the rest of 2013.